on the World Wide Web
Since 1995, when the first union web site was created, the labour movement has built the richest depository of union-related information and knowledge ever produced. But we can't use it. Worse, we can't even see it. It's scattered across a few thousand web sites existing on a World Wide Web which has some 80 million other sites. It's as if the greatest book about the labour movement - containing histories, policies, research data, educational material and more, in many languages - had all its pages scattered around the world. Looking at a few isolated pages of this great book is interesting, informative and often very useful, but you never get a sense of the whole work. The work, the labour of thousands of unionists around the world, needs to be seen to be appreciated and used.
The question addressed in this essay is: How can we make labour's information, knowledge and presence on the World Wide Web viewable and usable for labour education and research?
Below is a presentation of the ten-step logic behind the essay.
1. Labour's resources and presence on the Web has created a LabourWeb.
The LabourWeb is the entity created by combining all the web sites, databases, news services and other sources related to unions that are accessible via the World Wide Web.
2. We don't know what is on the LabourWeb or who is using it
We know there are many unionists and union sites on the Web. But we do not know who is using what Web resources for what purposes. We should.
3. The LabourWeb is being built everyday by people motivated by their local, organizational concerns.
By creating information and knowledge for their immediate needs unionists around the world are at the same time creating resources which can be used by the global labour community.
4. We can use the resources of the LabourWeb (information, knowledge and people) for labour's goals.
There is a vast amount of information and knowledge available on the LabourWeb which could be used to support labour's bargaining and political activities. This is especially important as more workers use the Web in their work to handle information or create knowledge. The LabourWeb is the most important resource ever created for labour research and education.
5. We need to see the LabourWeb, and participate in designing it, before we can use it.
The information and knowledge on the LabourWeb is so widely scattered through the Web that it cannot be easily seen as a whole or effectively searched by general, commercially-oriented search engines such as Google. It needs to be searched with labour-designed tools used by actively involved unionists.
6. The LabourWeb is a huge dynamic database of information, knowledge and people.
Considering the LabourWeb as one large database being updated everyday allows us to use existing database design and searching strategies to make it manageable and participative for unionists.
7. We need a participative web searcher (a tool which puts together web searching tools and people) to work with the LabourWeb.
A participative web searcher is my term for the combination of a World Wide Web search engine (like a Google) plus the community communication tools needed to allow people to participate in uncovering information, creating knowledge, and collaboratively learning on the Web.
8. SoliComm, a labour-designed web searcher, can be used to make the LabourWeb more visible and participative.
SoliComm (the Solidarity Communications system) has its own web-searching engine, file depository, email, web site hosting and computer conferencing systems - all especially designed for unionists involved in labour education and research.
9. SoliComm can be used today to help unionists use the LabourWeb for labour education and research.
SoliComm provides the only labour-specific search engine service on the Web. It is aimed at providing faster, more complete coverage of the LabourWeb than any of the general search engines can. It's a working system which we can use for practical matters now and for planning labour's role on the Web as it grows.
10. SoliComm could be used as a staging system and prototype for managing the next major evolution of the World Wide Web.
The next major stage of the World Wide Web - coming soon as the Web matures - will be organized as one huge discussion-like entity involving computers and people. It will be interactive, participative, intelligent and self-recording. It will be designed according to how humans think and talk. And, unless the labour movement considers what role it could play in this next stage, the mature Web will look like a union-free zone for most of its users. That could have serious implications for a labour movement which is trying to renew itself in the face of globalization and concentrated attacks on its existence.
This essay presents a few ideas on how we can see and use the LabourWeb in conjunction with SoliComm. It's designed to promote discussion of what could be.
Where is the union information I need now?
Labour activists are information-handlers. They are good at asking questions, getting information, and transforming that information into knowledge for union related work. At the local level they may need information on how to respond to a health and safety grievance involving a chemical they've never heard of. At the national union level they may be searching for information to support their arguments for sector-wide bargaining. If they are working on international issues they may need to know about migration patterns or the global ban on asbestos. And usually when they're looking for information, they need it now, not next week.
The sort of questions labour activists pose can be divided into two broad categories. They may be general (as in "I need to know if chemical ABC is carcinogenic") or they be labour-specific ("What unions have had a problem with chemical ABC and how have they addressed the problem?"). What these questions have in common is that they have probably been posed before and at least partially answered. What's more, given the rise of the World Wide Web in the past ten years, they have probably have been answered on the Web.
The general questions are the more easily handled. Just type the key words into a general Web search engine such as Google and you're likely to get thousands, if not millions, of "hits" (places on the web that include the key words). You probably would not have to go past the first page of hits to find out if the chemical was cancerous. The more difficult questions are the labour-specific ones. Filtering the thousands or millions of hits produced by a Google search to find out what labour thinks about the use of the chemical, and the collective bargaining response to its use in the workplace, is not easy. Unless you are a trained web-searcher (like a labour librarian, and sometimes not even then) you are likely to get lost in all the hits. You may never get to page 346 to find out that a union in South Africa had hired a consultant who produced a report on the use of the chemical, and that the union had used the report to bargain protections for its members. And there, right on the union's web site along with the report, was a model contract clause you could have used at the bargaining table.
But don't feel bad. It's not your fault you can't find relevant labour information on the Web easily. The truth is that the Web is being designed for commerce. Companies pay consultants and sometimes even the search engines themselves to get their information listed at the top of the list of hits. The union in South Africa would probably not have had the resources or the desire to do the same. So its experience was put on page 346, hundreds of pages away from the first page where the chemical company was announcing that everything was just fine.
The problem of information overload when searching the Web is not only related to the amount of hits that are displayed, but to the time it takes to find the relevant hits. I suppose you could sit in front of the computer and click on Next Page 346 times, but most labour activists are busy people and they need the information immediately, or they lose an opportunity to represent their members or labour organizations. Even labour people who are hired full time to research issues, develop training materials or create communication strategies cannot take the time to fully explore the responses to their questions. They're overworked and driven from crisis to crisis by circumstances. Page 346 is not only 346 pages away; it's two or three hours away.
There's another aspect to this overload of hits. The relevant hits - the ones of most interest to a unionist web searcher - may be scattered throughout the million or so hits which are found. It may be that in all that mess there are two hundred different instances of unions having trouble with chemical ABC. That many instances probably points to a pattern: the use of the chemical is causing concerns or health problems for many unionists. But because the hits are scattered widely over a vast number of hits the fact that there is a general problem shared by many unions is not highlighted. An important issue exists. But unions are left to address it in isolated ways and thereby are not learning from each other, sharing resource material, or banding together to start a campaign.
Now consider this: If you are a worker who has a question such as "Is asbestos dangerous?" and you visit a cybercafé to query Google, what happens when you don't see any union sites in the first few pages? (Most people don't even get past the initial ten hits). Well first of all, you probably won't get any information supplied by a union, so you won't be getting the full story. But secondly, and most importantly, you are left with the impression that unions don't care about the issue. The labour movement is not seen in the debate. It's invisible. And invisible quickly translates into: doesn't exist.
That is what we addressing in this essay: the disappearance of the labour movement as working people start to depend more and more on the Web for information, news, social interaction and education. Out of site; out of mind.
Data and Information for Knowledge-building
Working people are starting to depend more on the Web for a variety of reasons: they are using it as a tool at the workplace, as an information source for personal interests and entertainment, as a political instrument, as a way of socializing, and for education.
For unionists the important consideration is that many workplaces are becoming increasingly interconnected with the Web's resources. The boundary between a workplace and the outside world used to be defined physically (the factory supplying the just-in-time supplies was 20 kilometres away) or via communications such as the dial-up telephone or the fax machine. Either way there was a distinction between the workplace and the other place. Connections had to be temporarily created (as with the telephone) to be put into effect. No more. Millions of workplaces, even if they are not formally connected in any business-to-business or business-to-client sense, are connected to every other workplace on the Web, all the time. That means millions of working people, especially those in economically advanced nations, but also in the developing world, can interact with each other and the resources of the Web as part of their jobs. And they do.
The key to understanding more clearly what is happening to many workplaces today is the distinctions which exist between data, information and knowledge. Data are raw words and numbers. Think of telephone numbers and addresses displayed randomly, all in a mess. Information is created when data are organized into understandable categories, such as the columns in a telephone book. The potential for creating information is limited only by the amount of data and the possible combinations. Knowledge, on the other hand, is built when new data is found, manipulated into information and presented in ways which increases human understanding about something. The potential for knowledge-building is limited only by the human imagination. But here's the important point: what is one person's categorized, useful information can be another person's disorganized, useless data. The Web can produce many hits on various subjects and these can be combined to provide information, especially for companies, entrepreneurs, students and the like. But to a unionist looking for specific labour-related information all the hits the Web throws up are just noise, random data because they are not pertinent to the needs of the union. What is needed is a list of hits which pertain specifically to labour. Only when that labour-specific list is complied or displayed do the hits become useful information for the unionist. Then the information can be used to produce knowledge (such as how a particular group of workers can protect themselves against the effects of a chemical in their workplace.)
It is important to better understand this process of translating data to information and then to knowledge because it is changing how millions of us work. By producing the Internet (with its sub-component, the Web) and other information handling technologies such as the telephone, fax and photocopier, humankind is once again re-defining what it is to work. A hundred years ago work, for the most part, meant physical labour on the farm or the mine or the factory. Today it means, for millions of workers, data-collection, information-handling and knowledge-building. If a hundred years ago you had told a mine worker or a fisherman that people could make a living manipulating words or numbers on a screen you would have been thought crazy and ignored. If you communicate to workers today without understanding how they work electronically via the Web and other technologies you won't be thought crazy. You'll be ignored.
It's not just the economically-advanced countries which are affected by the advent of information-handling tools either. Developing countries are also affected. Some countries such as India and China are adopting information-handling and knowledge creation (in the form of call centres, computer programming firms, insurance data-handling companies and more) in order to create employment for sections of their workforces. But for the most part developing countries are being left out of the economic advances being spurred on by the Web and other information technologies. They lack the structures, such as the wide-spread use of credit cards or bank accounts, which are needed to take advantage of Internet-based commercial activity. And their educational frameworks are weak. The result is that they are not able to participate in the global Internet-based economy where transactions worth trillions of dollars are conducted everyday. That translates into lost opportunities for increasing employment (especially for young people), spending on health care and improving educational systems. Still, the opportunities for engaging workers in developing countries in Web or Internet-based activity is growing, especially because of the rapid spread of cellphone technology.
Education for l'earning a living
Understanding how to provide educational opportunities is the key to engaging workers in information-handling and knowledge-creation enterprises, plus provide potential for growth in the developing world.
Workers engaged in information-handling and knowledge building are what I call l'earners - people who earn a living by learning on the job. They manipulate words and numbers to produce information and knowledge to serve their enterprise and its clients. They do so within a range of learning opportunities which runs from none to infinite. For example, clerks who manipulate insurance data day in and day out do not have many opportunities for learning more about their work or the company. At the other end of the range architects can learn from their projects as long as they continue working. Most l'earners are employed between the two extremes. They learn a little or a lot as they go along. But what all l'earners have in common is the desire to learn and the ability to do so, sometimes in astonishingly rapid and effective ways.
A case in point. In 1980 there were maybe a few hundred microcomputers in the work world, if that. By 1990 there were tens of thousands of computers plugged into workplaces all over the world. The typewriter - remember typewriters? - disappeared in less than ten years. How did that happen? It happened because of two factors. First, people employed in information and knowledge-building workplaces have a tremendous desire to learn. It comes with the job. Because they manage information and knowledge for a living they are stimulated to want to continue learning. What's more, because they usually have a higher level of education, they have learned how to learn. Secondly, the vast amount of learning which occurs everyday is not the formal product of schools or company training programs. It is produced by people working out problems by themselves or in groups. It happens wherever people are, but especially in locales designed especially to address problems and produce solutions, such as workplaces. The microcomputer was introduced into thousands of workplaces by people who never took a computer course. They learned how to work microcomputers by themselves, or more often, by asking people around them. They will do the same in the future as new tools are introduced - whether those tools incrementally advance a technology (such as cellphones) or are tremendously disruptive (like microcomputers or the Web).
The growth of sectors of the workforce engaged in information-handling and knowledge-building has two major consequences which should concern union organizers. The first is that most l'earners need constant access to new learning opportunities in order to keep up with the changes in their work. They need life-long learning. The second is that millions of workers are frustrated because they are not provided opportunities for this learning. While it is true that the vast amount of learning that takes occurs in workplaces is informal peer-to-peer activity, workers understand that this informal learning could be built upon to produce even more advances which could improve their lives at the workplace and at home.
Some unions are taking advantage of this situation to motivate their activists, build greater membership loyalty to the union, recruit new members, and present unions as engaged in the new electronic world being created instead of being seen as leftovers from a fading industrialism. The British public employee union UNISON is an example. Encouraged by the country's Trade Union Congress' campaign for life-long learning, it is providing access to education for its members and potential members. It provides educational opportunities ranging from entry-level programs for workers wanting to return to formal learning all the way up to the provision of degrees in cooperation with Britain's Open University. The TUC and UNISON know that workers in economies which are increasingly based on information-handling and knowledge creation need access to life-long learning and just-in-time learning. They also know that unions will be seen by workers as influential actors in the new knowledge-based economies (and sectors of economies) if they negotiate educational and training provisions in collective agreements, create opportunities for working people to re-join the formal educational processes of the country, and lobby governments for the educational rights of people in the workplace. All this educational activity could be supported by a LabourWeb searcher which helps people find the uniquely labour oriented view of the subjects and issues they are studying.
Initiatives such as UNISON's are vital to the growth and development of the labour movement. However, considering that the vast amount of learning which occurs in workplaces is informal peer-to-peer, there is a need to augment labour's efforts by also supporting just-in-time learning opportunities.
Carefully planned and scheduled educational activity is a vital component in any educational endeavour. That's why one-day workshops, week-long schools and multi-year courses exist in the labour movement. But the overwhelming amount of education that occurs amongst working people is based on informal exchanges prompted by whatever is the issue of the day. Shop stewards, for example, are constantly confronted with new situations for which they need immediate answers. If they're fortunate, and their union has the resources, they can refer the question to a union expert. But even then they may not have the time to contact the union headquarters. They certainly don't have time to take a course. They don't even have time to figure out who to contact. They need to learn something immediately. They need Just-In-Time learning. JIT is the provision of opportunities for learning about a subject at the exact time the need arises. Traditionally labour activists would practice JIT by asking other workers or phoning other unions. But the advent of the World Wide Web has enhanced the possibilities for JIT tremendously. The challenge is to filter all the unorganized data the Web supplies into information which can be turned into useful knowledge.
Let's say you're a union president and you've just been asked to attend a workplace meeting with the employer on HIV AIDS. And let's say you know little about the problem, and even less about how unions are reacting to the pandemic. What can you do? Try this: go to Google and type in HIV / AIDS. How many hits do you get? A recent search displayed more than 25 million. Try a little more advanced search statement: HIV /AIDS UNIONS. That's better. Some unions are listed on the first page. But still, there are 5 million hits in total and they are mixed with results from the European Union, the American Civil Liberties Union, the Cambridge Students' Union, the Canadian Credit Union and so on. You could of course keep refining your search statements. Google supplies more than ten commands for creating a more focused search statement in its "Advanced" page. But 95 per cent of people rarely use them (probably scared by the word "advanced").
The point is that the Web is the greatest educational resource ever to be made available for the just-in-time learning needs of union activists, but it can be an overwhelming experience for most people. What is needed is a labour-specific web searcher which automatically narrows searches to unions or other labour organization sites and provides suggested links to other web resources.
But there's more. The impression should not be left that JIT on the LabourWeb is only about getting a focused display of hits. JIT labour learners need guidance. They need to know, for instance, the ten most relevant hits that labour people knowledgeable in the topic would suggest for beginners in the topic. They need canned ten-minute lessons on the subject they trying to learn about written by labour people. They need a forum of labour activists in which they can pose basic questions. They need a LabourWeb searcher.
Just in time learning on the LabourWeb could prove very important to an approach to unionism that many in the movement feel should be promoted more strongly. This approach, called "organizing the organized" emphasizes the constant organizing of union members to participate in their union's activities. It is in contrast to the "servicing" model where union representatives are given problems (such as an unfair disciplinary action) to solve with minimal involvement of the member. This leads to an insurance industry approach to unionism: members see the payment of union dues as insurance premiums which earn them protection if they ever get into trouble. An organizing the organized approach on the other hand is aimed at getting more people to be constantly active in their unions. Yes, it is unlikely that all members in a union will become active, but at least with an organizing the organized approach the circle of union activists will widen. What's more, unions that adhere to the approach position themselves for discovering new ways of involving their members that they probably would never have discovered. UNISON's programme for involving its members in educational activities (both union and academic) is an example of a very effective organizing the organized approach.
At the core of the approach is education. Opportunities for stimulating learning about unionism and the role of the union are constantly available: A promotion is granted by favouritism. Rumours of lay-offs are in the air. The company annual profit-loss statement is released. A new technology is introduced. The informal economy of the city has produced another death as street -sellers hawk their goods in traffic. The idea is to seize opportunities for organizing people into discussions about their working conditions and what they can do to improve them. However, just starting a discussion is useless unless it brings in new information which can be applied to the problem at hand. Anything else is just idle chatter which repeats old, probably discredited, arguments and failed solutions. In order to build the new knowledge which is necessary, new information is needed just in time for the discussion to have its most potent effect. A LabourWeb searcher which is available 24 hours a day from anywhere in the world could provide the labour-specific information that is needed for a particular group of workers to move from unfocused complaining to unionized action.
Just-In-Time learning about labour's views, struggles and solutions using technologies such as a LabourWeb searcher can be crucial to the development of unionism at the local and national level. But it can be equally important to the development of the international labour movement, especially given the expanding effects of globalization. The current round of globalization is based on the intermingling of economies, the creation of new international governing bodies such as the World Trade Organization and the International Criminal Court, plus the strengthening of long-standing establishments such as the International Labour Organization (which defines basic international labour standards). A realignment of how we trade and govern ourselves internationally is taking place. And it's all relatively new. The start of the current round of globalization can be arguably set to about 1989 with the collapse of the Soviet Union and the introduction of the microcomputer connected to the Internet and the Web. The result is that labour organizations all around the world have had to grapple with issues such as free and fair trade, intellectual property rights, migration and global governance. Not surprisingly, these organizations vary in their views about globalization, its effects and potential responses. There is no one labour view about globalization. What the American labour movement thinks about free trade may not be what the Tanzanian movement thinks. There is a need for a global labour discussion about the issues - a discussion which involves grass root unionists, national leaders, and international labour activists. And this discussion cannot be relegated to the occasional international conference. It needs to be conducted at the local and national level when opportunities promoted by yet another effect of the phenomena. What's more, it needs to be nourished with labour-specific information which can lead to the knowledge necessary for promoting a globlalization tied to social as well as economic advances. Some of this information can be provided by the international labour organizations such as the ITUC and the Global Unions. But they are under-resourced, under-staffed, and overloaded. What is needed is a marshalling of all the information that exists about labour's reactions to globalization, especially the information on the Web, in coherent and easily searchable ways to help unionists in their discussions, when they need it.
The fact is that labour is having a grassroots, local, national and international discussion about globalization. But it's so disordered that at the international level it is impossible to see, never mind understand. It's scattered across thousands of web sites and a few email lists. It's being held in many languages. And it ranges across an immense list of topics. A LabourWeb searcher however, would be able to bring focus to the discussion by making visible the content of all the labour web sites in the world in easily searched and categorized ways with simultaneous access to labour people knowledgeable about the topics.
Online news, learning and research
But it's not only the discussion about globalization which would be served by the use of a LabourWeb searcher. The searcher could also be used for finding labour news, supporting online (via computer communication) learning and labour research.
The web searcher would be used quite effectively for labour journalism and news reporting. For example, news could be searched according to sector, topic, region or other criteria. The ITUC, the Global Union Federations and the national labour federations produce news every day. But, like much of the news on the Web, it can be difficult to assimilate because there's so much of it scattered amongst so many sites and it's impossible to search as a single entity. Meanwhile, the labour news service LabourStart has been remarkably successful in providing labour news using a combination of computer tools and labour activists. LabourStart deserves to supported by more labour organizations. It should also be augmented by the parallel development of a LabourWeb searcher which could search all the text of union-created news items.
Labour educators could use the the LabourWeb searcher to look for material other educators have produced, find information for developing course materials, and build online activities into their courses. This last possibility has the potential to greatly expand the reach of labour education. The ideal situation in labour education is to have the participants all in one room so they can learn through peer work and discussion. But distance education can enhance these activities by allowing the participants to work in online computer conferences before the face-to-face activity and provide group support afterwards. This sort of blended distance learning (which blends face-to-face with online education) allows labour educators to design programs where participants can stay in their workplaces and attend occasional residential activities. Online learning has been proven to provide effective labour education which promotes community and collaborative work. A LabourWeb searcher could support its growth.
And, of course, unionists would also find a LabourWeb searcher useful for research. The LabourWeb can be considered a huge database of labour-related information being used every day by thousands of labour activists. And it's growing. If it could be questioned in easy but effective ways it could prove to be the greatest aide to labour researchers ever produced. It could prove especially effective for unionists in developing countries. For example, while microcomputers are widespread amongst central labour federations in the developing world (even if these computers are old and underpowered) individual unions do not have much access to the technology. But cellphone use is expanding at a fantastic rate even in the poorest regions of the world. Workers researching a topic, such as a new chemical being used in the plant, or the laws related to child labour, could use their cellphones to send requests for information to the database and the people using it, and then hang up. Later, a text message could be sent providing the requested information or saying that some background material could be mailed. Technology does not have to be designed only to serve the needs of those in the economically advanced countries. It could be designed to serve the whole labour community, even in its poorest regions.
The first step though is to start making the LabourWeb more visible on the Web so people can see its potential for focused labour research and education.
There is a series of books for children called "Where's Waldo?" about a tiny cartoon character who is lost in a big picture of animals, cars, houses and whatever. The object is to find him - and he's not easy to find. He's almost invisible in all the stuff. Labour on the Web these days is a little like Waldo. It's almost invisible in all the stuff.
Partly this is because the search engines are designed for commercial purposes. Type "lay-offs"in Google and you'll get lots of advice about how not to get laid off, or what to do after a lay-off, by booksellers, newspapers and employment companies. You'll even get advice on how to implement lay-offs if you're a boss. But nothing from a union. Try a few other searches yourself: type in "poor wages", "firings", "employee health plans" "health and safety" - the kind of keywords a person with a problem at work might enter. If you are a unionist you will become quickly discouraged. Nowhere in the first two or three pages (which are more than most people view) can you find a union offering to give advice or help. The only conclusion a worker surfing the web can come to is that unions are not involved in these issues and don't care.
This invisibility is partly a product of the number of hits general keywords such as "lay-offs"produce and the importance general search engines allocate to unions. But it's also because unions have exclusively concentrated on building their own web sites. And the central labour federations, which are responsible for promoting labour in societies, do not have the resources or the inclination to become actors on the wider Web. So the Web, as portrayed by the search engines, is left to be seen as a non-union environment.
It gets worse.
The single biggest phenomena to hit the Web, second only to its commercialization in the 1990s, is the growth of social networking. Millions of people are participating in the Web to find friends, share pictures, recommend interesting links, talk about issues, provide news, and more. If you define social movement as "people moved to do something socially" it is the most important social movement in the history of humankind. And through all this the labour "movement" is nowhere to be found.
Don't believe me? Try an experiment. Go the site called del.icio.us which describes itself as a social bookmarks manager. It is used by millions of people to share their lists of important Web links. Enter: "poor wages". Not one union website is linked. (When I did this search the top hit was a site by the libertarian economist Thomas DeLorenzo entitled "How 'Sweatshops' help the poor".) Enter the word "unions". The result is a little better. The Australian Council of Unions is linked to an article on how to join a union. The LabourStart site is marked. And the Global Unions site is mentioned on page two. But the preponderance of hits involves anti-union newspapers articles and websites such as "The Evil of Swedish Labour Unions", "Death to Unions" and unrelated hits such as "Same-sex unions". Try another experiment. The Web has thousands of discussion groups aimed at topics ranging from Apple picking to Zebra breeding. Try to find one which discusses the international labour movement and its issues. You can't. It doesn't exist. There is no central place on the Web for a worker in Africa or anywhere else to ask a question like "I've heard of something called the ITUC - what is it?". This is not just an information-providing issue. It's related to the visibility of labour on the Web.
All this has an obvious implication: if the labour movement is invisible on the Web, the next generation of workers will not see it as helpful or relevant and this will contribute to a further decline in labour membership numbers and therefore our power to represent the interests of working people nationally and internationally. That is a serious consequence for anybody who believes that unions are still the best instruments working people have to protect themselves and improve their working conditions.
An even more serious consequence of the invisibility of the labour movement is that its political views are also lost amongst all the noise on the Web. Since its inception the labour movement has been a major source of politically left views on the conduct of government, the role of employers, the social and educational needs of workers, the end-purpose of economic activity and much more. If labour's voice is lost or outshouted on the Web, society loses a major proponent of a way of thinking that says the goal of the actors in economic activity - such as governments and corporations - is to provide full employment, adequate education, decent social protection and time to enjoy life.
The Web is having a civilization-changing influence on how humankind thinks and learns as well as how it manages the creation, stocking and manipulation of knowledge. If Labour wants to play a part in this it has to start making itself more visible to its members, potential members and the public. One instrument, in the whole array of instruments which will be needed to do this, is the development of a tool which highlights labour on the Web. We need a LabourWeb searcher. Read on.
Or not. If you've gotten this far you've got the basic idea: There's a part of the World Wide Web which consists of all the labour-related sites, mailing lists and databases which we can call the LabourWeb. We can take advantage of this LabourWeb by developing a labour-specific search engine tied to communication tools such as computer conferencing, email, site hosting, radio and Internet telephoning. What follows is a more detailed description of the technology - how it is being designed and how it could be designed. You could stop here and still be involved in the project by using the SoliComm web searcher (at www.solicomm.net). But if you are interested in how SoliComm will be designed with the help of unionists like you, read on.
What is a labour designed technology?
But before we can discuss the development of a labour-designed technology such as a LabourWeb searcher we have to define what a technology is. This is important because the prevailing definition is based on an engineering orientation which minimizes the role of people. Our job as technology organizers (for that is what we are if we design technologies within the labour movement) is to do the opposite: it is to maximize the role of people and their capabilities.
Here is an engineer's definition of a technology: a technology is a tool (hardware, software or mental) which is used to solve problems. The great difficulty with this definition is that it excludes people. It is as if technologies were not to be used by or for people. By omitting men and women in the process of designing and using tools the definition excludes public debate, corporate interests, governmental action, cultural imperatives, psychological orientations, fun, love, sex and more. This is the definition adhered to by the designers of Google as they try to build a search engine which can act on its own without the intervention of people.
Here is a more people-inclusive definition of technology: A technology is a tool (hardware, software or mental) which is used by groups of people to solve problems. Technologies become technologies (as opposed to scientific entities) once they come out of the lab into use by people. And they are used by groups of people because that is how we organise ourselves (technologies are rarely designed for one person.) A technology is a tool (such as a computer system) plus people. This way of thinking about technology is especially important for unionists because unions get their power when people work together in groups to improve their employment conditions. If we are going to design a technology, such as a LabourWeb searcher, we should do it with the idea of enhancing the power of people to work in groups.
The labour news service, LabourStart, is a concrete example of how a technology is designed along labour-oriented principles. Its designer, Eric Lee, would be the first to acknowledge that LabourStart is not the collection of programming tools he has written. Nor is it an attempt to write programs which automatically go out into the World Wide Web to find labour-related news stories. Instead, its core function is to enable a group of labour activists from many parts of the world, working in many languages, to pick news stories that they think the international labour community would find interesting and useful. LabourStart is an example of the technology organizer's definition of technology: a tool plus people.
Another example: in 1983, while I was director of the computer department of the Canadian Union of Public Employees (CUPE), I introduced the microcomputer and the country's first Local Area Network to the union as tools for accounting, word processing, databases and other applications. At the same time I designed a communications system (called SoliNet) to run in parallel with the tools. This allowed people to discuss how the tools were used and the problems associated with their use such as health and safety concerns. SoliNet became the first union-owned computer conferencing and email system in the world. It was opened to the whole labour movement, national and international, and spawned a number of important projects such as the first online labour education courses and the first home for LabourStart.
Tools plus people; that's the labour way of organizing technology. It is also quickly becoming the way to create powerful software applications which labour can use to design technologies for itself. The "free, open source software" movement is producing a radical change in how humankind interacts with computer technology. Open source software is software whose "source code" - the underlying instructions telling the computer what to do - is made available to the public. Anybody can copy it, change it and redistribute it without paying fees or royalties. Open source programs develop though community collaboration involving thousands of volunteer programmers as well as very large companies such as Sun and IBM. It is the largest collaborative learning phenomena is the history of the world. And, not surprisingly given its historical consequence, the free, open source movement is not without its controversies. For example, the word "free" which is often used alone or in conjunction with the words "open source" does not refer only to lack of usage charges, but to "freedom". That raises important political, social and economic questions and distinctions (for example, the system I designed in the early 1980's, SoliNet, was open source, but not free, because we had to pay to license the underlying program). However, as important as the discussion is, we don't have the space to explore it more fully here.
What is important to understand about the open source movement is that labour organizations can use its products for the production of computer systems and software distribution programs which would have cost millions of dollars just a few years ago. For example, I provide the participants in my computer technology courses a CD which includes software for word-processing, spreadsheets, databases, presentations, email, web-browsing, web site creation and more, all free of charge. Many of the programs we use in our central computer systems, such as our email, web searching and database services, which five years ago would have cost tens of thousands of dollars, are open source. These sorts of programs allow us to leverage our hardware resources to enable technology projects that in the 1990s would have been considered airy dreams. The challenge is to marshal our human and technical resources to take advantage of the opportunities the free, open source movement is providing in order to create technologies which serve labour's goals - technologies such as a LabourWeb searcher.
A Labour Search Engine+
A LabourWeb searcher can be considered a search engine plus. It has the tools which enable web sites to be searched (rather like Google) plus the communication systems to augment the searches with human participation. If somebody searches for "job security" for instance, the hits are displayed à la Google, but at the same time the person is provided an opportunity to explore the issue in networks of real live unionists. Hence the descriptor: participative Web searcher. What follows is a description of a LabourWeb searcher divided in two sections: the search engine component and the communication systems.
The search engine component
A Web search engine is an information retrieval system which searches for information on the Web. The engine takes one or more key words a user types into a box on a web page (a query) and displays "hits" - places on the Web where the words can be found. These hits are accompanied by the URLs (the web addresses where the material can be found). The hits which are displayed are ordered by criteria set by the search engine company. For example, Google uses its system called PageRank to determine where on the list of hits a page should be put. It assumes that pages with the most links from other pages are more important and should rise to the top. If you enter "job security" Google will provide you a list of pages which have the term and order the pages like this: : hit 1 (a page which includes 500 links to it), hit 2 (300 links). . . hit 3,000 (2 links). In other words if you want a page to be put at the top of a Google search result make sure it has a lot of links to it. (It's a little more complicated than that, but that's the basic idea.) Some search engines, not Google, will mix hits found in the normal way with hits paid for by clients who want top placement for their pages. Google will place an advertiser's page at the top, in a shaded area separate from the other hits.
Google's PageRank system is why when you type in "union" you see "Union College, Schenectady, NY" (which is unrelated to the labour movement) at the top of the list. The Union College home page (the first page on its web site) has more links to it than the AFL-CIO's (the US central labour federation) which is fifth on the list. Labour's major international federation - the International Trade Union Congress - is nowhere to be seen. (I gave up searching for it after the first ten pages). If every labour site in the world included a link to the ITUC's home page it would be at the top of the list.
Search engines consist of three main components: 1) a program which searches for keywords on the Web (called the "crawler"), 2) a index of all the keywords which were found, and 3) the query processor (which is the program you interact with when you enter a query and see a list of found links). When you use Google you are working with its query processor. It goes to its massive index of keywords, finds the words and then displays them, along with the URL where the words can be found. In the background the crawler searchers the Web or parts of it every so often to refresh its list of keywords. Simple, no?
Well, it might be if the Web consisted of a few sites. But there are more than 80 million sites in the world and the number is increasing every day. Crawling, indexing and displaying results from all those sites adds up to a massive task. It also results in phenomena such as the labour movement disappearing amongst the millions of sites. The answer is to build a search engine aimed specifically at labour sites.
There are two basic kinds of search engines: general (also called generic or horizontal) and specialized (also called topical or vertical). Google is a general search engine. SoliComm is a specialized one. The major difference is that general search engines are aimed at the whole web. Specialized engines are aimed at a specific number of sites chosen by a person or team. When it began SoliComm was aimed at 300 of the largest labour websites in the world with a total of 1 million web pages. Nobody is sure how many labour websites there are, but let's guess at 1,000. The goal of SoliComm's search engine is to provide better search results from those 1,000 than Google or any other general search engine can.
A specialized search engine can provide better results in its field than a general one for a number of reasons. First it automatically supplies the context in which the query is posed. If you type the word privatisation in the SoliComm query box your search is automatically focused on labour union web sites. The results are not mixed up with hits from the European Union, credit unions, and university student unions.
Secondly, a specialized search engine can probe deeper into sites to look at all their pages. Google, because it is a general search engine aimed at indexing the whole Web, necessarily has to restrict the number of pages it looks at. Normally that means about 90% of the site is indexed, which is impressive. But what about the other 10%? It could include important content which is being missed. This content adds up to a huge amount of information which is not accessible to unionists trying to search the LabourWeb. A specialized search engine, however, can be focused on a particular set of sites (instead of the whole Web) and therefore provide more comprehensive results. SoliComm is designed to search 100% of the sites in its list.
A specialized engine could also provide better results than the general search engines by providing access to the labour part of what has become known as the "deep web". The deep web is the information general search engines cannot access. It includes databases which demand a username/password combination before their data can be displayed and web pages which are presented only after a user has entered a query in the website's search box. General search engines cannot plug in usernames and passwords. They cannot put in queries to call up web pages otherwise inaccessible.
We are not sure how much information is contained in labour's section of the deep web, but it is probably substantial. A specialized search engine could make this information accessible to union researchers and educators in two ways. First, labour web editors (sometimes called webmasters) could supply an index of all the web pages in their sites which general search engines do not have access to. SoliComm, or some other LabourWeb searcher, could then include these indexes in its central index and thereby make the pages available to unionists. Or, if the union web sites are using databases compatible with the specialized search engine it could be trained to work with the databases directly. Either way, the union web editors are more likely to allow this sort of activity because they know that by doing so they are not contributing to the commercial interests of Google and the like, but instead to a search engine controlled by the labour movement for its purposes.
The cooperation of labour web editors in the development of a LabourWeb searcher could be extended in other ways. For example, we could begin a process of what I term site-bonding. This is when sites begin to include elements from other sites. The news feed provided by LabourStart to union sites around the world is a clear example of site-bonding. The sites which carry the feed (the stream of news stories) from LabourStart have bonded to it. The result is mutually advantageous: the union sites get regular updates of labour news and LabourStart gets access to more readers.
Site bonding for a LabourWeb searcher could involve a number of strategies. Most obvious would be insertion of search boxes on union web sites. Google does this; SoliComm, or another LabourWeb searcher, could do the same. A unionist entering keywords into the SoliComm box on her site in Sri Lanka would be provided with focussed information from union sites all over the world. The opportunities for just-in-time labour research and education would be significantly increased.
Another bonding strategy would be to create an automatically updated and searchable list of labour sites, classified by particular categories, which web editors could include on their sites. The Web searcher could automatically group unions according to categories such as employment sectors, job classifications, regions and particular issues such as privatization. It would make that group searchable on its own, as a subset of the LabourWeb. And finally it would put the list on all the union web sites subscribed to it. The list would be updated every time a web editor somewhere applied a new category to her web site. Members visiting their union's site would see a list of categorized links which would be updated every time a new union was added to the category. In this way unionists could search within their sector or for particular issues. The key would be to have web editors use a web page element called a "meta tag" to help categorize their sites. A meta tag is a description such as "public employees" that a web editor puts at the top part of their pages, in a hidden section. They were popular a few years ago, but then some commercial elements started misusing them in order to fool search engines such as Google. (For example, a tag like "education" would point to a gambling site). However, union web editors could be trusted to not maliciously misdirect people with fake meta tags. The result would be a categorized view of all the labour sites in the world, automatically updated by union web editors working on their own sites. Sites on the LabourWeb could have a live links page connected to every other labour organization site in the world. The LabourWeb would become much more visible and therefore more useful.
The categorizing of the LabourWeb by union web editors could be augmented by computer power. When SoliComm searches the labour sites in its target list it creates a table of keywords. This table could be used to create categories and links to sites which include resources connected to those categories. The categories could be created by analyzing the number of keywords. If there were substantial instances of a particular keyword on more than 25 sites on the LabourWeb then it's a likely possibility that it represents a category labour people are interested in. Then the sites from where the keywords came from could be analyzed to see which of them had a high percentage of a particular keyword and the site could be put into a specific category in an index. For example, if "migration"is found as a keyword in more than 25 sites it would be declared a LabourWeb keyword. Then the sites which contain a high percentage of the keyword word would be listed in a category called migration. This is not the place to exactly define the strategies and algorithms by which LabourWeb sites could be tagged and analyzed. However, the idea should be clear by now: by recruiting union web editors to categorize their sites and augment that human endeavour with computing power we could end up with an organized and usable view of the LabourWeb.
Searching the Whole Wide Web
Still, as important as it is to make the LabourWeb more visible, we cannot forget that unionists are also searching the whole Web for information which could help them in their union work. We need to also consider how people in the labour movement look at and use the Web in its entirety. Two strategies could be adopted: we could create a live links pages of the whole Web, and we could use a Web meta search engine.
Most union web sites have a page where they suggest exterior Web links. These pages make up an important resource because union web editors have chosen sites they think would be valuable to their members. These links would be useful to other unionists as well. By collecting all the links on these pages, and of course eliminating the duplicates, we could produce a directory of the Web created by union web editors. SoliComm, or another LabourWeb searcher, could be told to periodically check the union links pages for additions or deletions and update the global list. The updated list could then be transmitted to all subscribing union sites. In this way a particularly labour view of the Whole Wide Web would be provided and union sites subscribing to the service would have a constantly updated resource for their members.
However, even a global labour links page would not be able to provide a comprehensive search of the whole Web for unionists. To do this we need a meta search engine.
A meta search engine is one which works on top of a number of individual search engines. It collects a query from a user and then sends that query to all the search engines it works with. For example, SoliComm has built into it a meta search engine called Ixquick. When you enter keywords into the box on SoliComm which says "Search all the Web" they are passed to Ixquick, which in turn sends them to 12 search engines such as Yahoo and MSN. The displayed hits are the ones which were found from searching all 12 search engines.
The first advantage to SoliComm's use of a meta search engine is that 12 engines are better than one for producing results. The second advantage is related to privacy. When you use a search engine like Google every one of your searches is captured and put into a file about you. These searches are never deleted. They are kept in order to build up a profile on you so you can be targeted more effectively by advertisers. The profile becomes even more detailed if you use a related service offered by the company, such as Gmail, Google's email service. In Gmail all the words in your emails are searched in order to aim particular ads at you. If you have been using Google and Gmail for a few years, Google has a lot of information about you, more than you know. And it's not just Google that does this, most of the other search engines operate in the same way because that's how they make money. So if you are a unionist searching on keywords such as "strikes" or "union organizing" you should know that you're being watched. What's more governments are forcing the search engines, even Google, to secretly provide the information they have on users under the guise of anti-terrorism activity.
Ixquick - the meta search engine built into SoliComm - works differently than most general search engines. It does not store its users' datastreams. It destroys the personal data of its users after 48 hours. So even if a government were to force it to provide the data it has on its users, at most there would be only 48 hours worth of activity to be reported.
Ixquick is just a temporary solution though. In the next stage of its development SoliComm will have its own, custom-created, meta-search engine. The SoliComm meta engine will search multiple search engines and protect privacy, as does Ixquick, but it will also provide a unique just-in-time labour research and education notification facility.
When you use a general search engine like Google or Yahoo you are presented with ads in a side column: "sponsored results" they are called, as if they were displayed for altruistic reasons not profit-making. These ads are very effective for the search companies because they are triggered by words entered by the users themselves. If the user searches for "car", ads from automobile companies are displayed. Advertisers love this formula because they can focus their advertising campaigns at people most likely to respond.
A LabourWeb searcher could do the same, except without the profit motive. Labour-created notices could be displayed in reaction to keywords entered by users. For example, if a user searched for "asbestos" a side column could display notices related to the search such as the ICEM's campaign against the product, news stories about compensation awards, policy statements from central labour federations, and so on. If somebody typed in "poverty" a side notice could provide a link to the ITUC's anti-poverty campaign. The idea would be to guide unionists searching the Web to labour resources, campaigns and other sources of information. The notification facility, which would also be used in the the SoliComm's labour-specific search engine, would provide a much more labour-oriented Web search experience for unionists and the general public.
Similar to the notification facility is the related keyword function. In Google, and other search engines, when a user enters a keyword a suggested list of related keywords is displayed. For example, if you search "union" on Google you are presented with the following list of related searches: union sql, civil war union, union bank, union c++, union oracle, union jobs, union leader, confederate. In this list only "union jobs" is related to the labour movement. What a nice message that sends to union members looking for information about unions: unions don't exist to represent their members, they exist to provide jobs for their leaders and staff.
A LabourWeb searcher, like SoliComm, could be designed to produce related keywords that are more clearly focused on the labour movement. A search for "unions" could produce a list of keywords such as: the name of the central labour federation in the user's country, the International Trade Union Confederation (the ITUC), collective bargaining, freedom of association, international labour standards, and so on. The list of keywords related to particular searches could be devised by panels of subject experts in the movement.
A related keywords facility is important because 95% of search engine users do not use advanced search commands. They enter a word or two. If they don't find something relevant in the first page, they try other words or just give up. The advent of search engines has produced an historic growth in information searching, but the vast amount of this searching is at a very primary level. People need assistance in their searching from automated procedures such as related keywords and ready-made search statements prepared by subject experts.
But that is not nearly enough. In terms of labour movement organizing it doesn't matter how many automated procedures we develop to assist searchers and provide clearer views of the LabourWeb if we do not, at the same time, build in ways for people to interact, to help co-unionists with their searches, to communicate with people in the movement. A labour technology is not just a tool, like a search engine, but also the people who use and design it.
The Communications Components
Two phenomena are affecting the World Wide Web as we head into the 21st century: search and social action. The former is changing how we think and find information. The latter is re-defining how masses of people interact. Both are provoking radically new ways of creating knowledge. These two phenomena have been growing up separately, but they are bound to start converging. The labour movement can wait until they do and once again react to how the technology is designed by others, or we can be part of the design. SoliComm is a first attempt at being part of the design of the Web so we can influence the technology as it matures and adjust it towards our goals.
Here's what's happening: Search engines, such as Google and Yahoo, are providing very powerful ways to find information on the Web. At the same time social action networks such as Fickr and Del.icio.us are providing web-based centres for people to share resources (such as pictures and links). Notice that that the two phenomena - searching and social action networks - are not connected. When you search on Google you do not have access to a community of people (like the community formed by Del.icio.us or Flickr). And when you use the community resource-sharing centres you are boxed into the system, with no search facility aimed at the vaster resources on the Web. Search is being treated as search; community sharing as community sharing.
SoliComm addresses both phenomena in an integrated fashion. It has a search component which provides a way to work with the LabourWeb (as well as whole Web). And it has the communication tools to enable communities to be an integral part of the search experience. When you search on SoliComm you have, at the same time, in the same system, access to networks of communities which are part of your information gathering.
The section above described SoliComm's search components. The section below looks briefly at its community communication tools: email, web site hosting, file sharing and computer conferencing.
SoliComm members are automatically given an email account. In these days of ubiquitous email (at least in the economically advanced countries) providing an email service may seem superfluous. But thousands of unionists in developing countries and even the economically-developed countries still do not have email accounts. We can wait until they all subscribe to advertising services like Microsoft's Hotmail or we can take advantage of an historic moment to create a global communications community by encouraging union members to create and use their own email system. There are about 1 billion users on the Web today. By 2012 the number is expected to double to 2 billion. There are many more people about to join the Web and many of them will be unionists who need email services. Beyond this opportunity though there are other reasons for providing email services: SoliComm provides an environment free of corporatist and sexist advertising. It offers encrypted email so unionists can talk to each other without fear of governmental intrusion. And it provides a central directory. This directory is small right now, limited to SoliComm members. But it could be used as a global email directory of unionists by including all email addresses of unions in the world. Since this directory would be situated in a protected environment commercial interests such as spammers would not have access to it. However it could be used by global labour organizations, such as the ITUC, for mass mailings related campaigns and information distribution.
A related service could be a universal spam-protected union email network. Any email address in SoliComm's directory or any union domain (the last part of an email address) could be tagged as allowable through union firewalls (network protection systems).
Email in the SoliComm system comes in three modes. It has a Web-based system which works like Hotmail and similar systems. It has what is know as a POP system which allows computer based programs (called "clients") to retrieve mail, read and write while not connected, and then send created emails when the connection is activated. The POP system works with programs such as Thunderbird, an open source equivalent of Outlook (except better). And finally it has an IMAP facility. This allows you to use programs such as Thunderbird to work with your email online so you can access your email from anywhere, but with the capabilities of a a normal email client program. It's rather like having a combination of Hotmail and Outlook, except that it's based on open source technologies.
SoliComm also has a website creation and hosting service. If we are going to encourage the use of the Web for labour education we need to provide labour educators with easy to produce web sites which they can use as part of their course offerings. Secondly, if we going to promote the use of the Web by unions we have a responsibility to assist unions which do not have the money or technical resources to create their own web sites. SoliComm provides free web site creation and hosting for unions in developing countries.
File sharing is also part of the current SoliComm system. Anybody can create a folder into which other SoliComm members can transfer and download files. Online tutors can create private folders to which only their course participant have access.
SoliComm also has a computer conferencing component. (Conferences are often referred to as "forums"). This allows groups to work in conferences which are open to the whole SoliComm community or restricted to only the members of the group.
Computer conferencing is essential for online labour education. Email is great for one-to-one communication and one-to-many (think spammers), but it is not suited for group interaction. For online labour education to be effective, messages have to be gathered into a common area so that they can be seen by the whole group, as the work of the group. This is how solidarity is developed online. And it is group solidarity that is at the core of labour education because that is where unions get their power. Emails scattered throughout an in-box that is littered with spam and full of messages from friends and co-workers cannot be easily grouped (by normal email users) and therefore the group collaborative work spirit which is essential to labour education is very difficult to generate. In conferences all the members of the group can see the collaborative work they are producing and solidarity becomes part of the process.
Computer conferencing, although it has not been seen as such until now, is also essential to searching on the Web. If you cannot find what you are looking for on Google what do you do? If you are a relatively sophisticated Web user you might look for a forum which includes people that may have the information you are looking for. But that means leaving the system, finding a possibly relevant forum, subscribing to the forum, entering a message, and then waiting for an answer. Most people don't even get to the second page of a Google search, never mind go through this convoluted process. The result is that most users simply give up.
However if, as they were searching in a labour-specific search engine, a side column could indicate conferences on the topic, they could go immediately to that conference. They would not have to leave the system to go elsewhere to pose their question. In the conference there would be real people available to help. In this way the collective information and wisdom of the community- in our case the labour community - would become part of the search process.
There are bits and pieces of this strategy apparent on the Web. People can use Google. They can join forums. But the two modes have yet to converge in any substantial way. SoliComm is the first specialized search engine/communication system to explore this convergence. It could lead to a new possibilities for how unionists search the Web.
One of the possibilities is to extend the system's capabilities to help labour people interact not only to explore the LabourWeb (with for example, live labour links pages and conferences on the topic) but the World Wide Web, all of it. The challenge would be to index the Web according to the interests and needs of people in the labour movement. The index would include not just labour-produced resources, but also point to other resources which could be useful for unionists in their work. Producing this index and keeping it up to date would be an immense task. But it would be possible if we take the example being set by social networks such as Flickr and Del.icio.us and extend the models they are developing.
Flickr (picture-sharing) and Del.icio.us (web link-sharing) ask their users to to label items on their sites with tags (descriptors of the items). The process of applying these tags, (called, appropriately enough, "tagging") is leading to a phenomena known as "folksonomy" where thousands of people tag items such as articles, pictures and music files. Eventually, the cloud of tags becomes a way for a community to find stuff. It's an anarchic way of doing things because everybody can add whatever tags they want, but eventually the community works out what tags are most effective to describe particular items and starts categorizing them. The tags become, in effect, a shared vocabulary by which the community describes the material it is interested in.
Current folksonomy systems such as Flickr are limited in that the content being tagged has to be one one system, put there either by the site organizers or the people using the system. A more comprehensive strategy would be to start tagging material that is on the whole Web, not just an individual system. Consider this: a unionist reading about child care on a web page somewhere decides that the page includes information would be of interest to other unionists. She right clicks on the mouse and chooses "Send to SoliComm". This opens a form in which she can tag the page, with keywords and a descriptor sentence, and then send the address of the page to SoliComm. The information about the page is entered in the SoliComm Web directory in an existing category or an "other" section from which new categories can be developed. The power of thousands of people in the labour movement looking at all the information on the Web, including that part I've labelled the Labour Web, would be harnessed to produce a labour-specific view of the Web. The possibilities for making the Web even more effective for labour people would be significant. But the effects of an operation like this would go beyond just the labour movement. Thousands of people on the political left searching and categorizing information on the Web would create a considerable body of knowledge which could be applied to the improvement of the living and working conditions of humankind.
By encouraging unionists and others to use a LabourWeb server such as SoliComm - with both its labour-specific and general search engines- we could also start building sophisticated analyses of their needs in order to provide better targeted union resources. Without identifying individual users we could track how people use the systems: what issues are being searched? what information is being requested? what regions are more interested in particular topics? What information is being tagged? Then we could begin focussing our Web-based information provision to actual user needs instead of what we think they are or should be doing on the Web. The key would be to build in stringent privacy protections. No individual's data traffic would be recorded. Only the aggregate figures involving the searches made on the system would be collected. This would be a way of protecting people's privacy. But also, the very context of the searches minimizes the danger of gross privacy. People are unlikely to use a specialized labour search engine to look for personally sensitive information such as medicines or particular entertainments.
All that we have been discussing above is possible today with tools currently available. We could have a LabourWeb which is more effectively searchable for unionists. We could have the Whole Wide Web indexed according to the interests, needs and political orientation of labour activists. We could make the Web a more useful instrument for the labour movement. But more importantly, by doing so, we would position the movement to be a major player in the next evolutionary step in the maturing of the Web.
The Mature Web
The next stage in the growth of the Web is to make it work more like computer programs do.
In this stage you will be able to pose natural-language questions and receive answers which will provide the information you want, plus the information you need, but didn't know to ask for. More importantly you will able to activate computer programs which will perform sophisticated tasks on your behalf using all the resources of the Web. And this has nothing to do with the kind of super-powerful computers you see on Star Trek or Star Wars. It just means a different way of coding information on the Web.
It's called the Semantic Web. Its lead designer is the inventor of the Web, Tim Berners-Lee, which means it's not just idle chatter or commercially-oriented hype (such as the currently fashionable "Web 2.0").
Web pages today - even the most advanced - are really just files of information which can be read using browsers like Internet Explorer or its open source equivalent, Firefox. The browsers are used like low-level readers: here is a headline, here is a link, play this music, show this video. The pages are not actionable by computers. Computers can't read the pages for instructions on how to perform acts such going out into the Web to create an original report on child labour in the textile sector. As the Web matures into what Berners-Lee calls the Semantic Web, information on the pages will become computer processable by browsers or agents which will find, share, combine and report on information more easily. The agents, (think of more intelligent browsers) will act like computer programs which can follow instructions to automatically perform tasks. A computer user in a country could provide an agent with a task, previously undefined, such as this: go find all the collective agreements expiring in the next year or so which mention chemical hazards, calculate the number of workers involved, find out the affiliation of their unions to national federations, and report back with all the email addresses of the presidents of the unions involved. By tomorrow. The report could be used to coordinate a national health and safety campaign tied to current contract negotiations. The individual efforts of the unions would be made more powerful because they were part of a wider campaign.
Of course the Semantic Web will be used for much more than labour related activity. Web users will be able to do things such as ask their agents to book medical appointments ("find a doctor in the city who specializes in migraine headaches and has office hours on Mondays or Tuesday and bills according to my health care plan, then book an appointment if there's no conflict with my calendar".)
It's called the Semantic Web because computers will understand the semantics, the meaning, of a document instead of just interpreting a series of characters. And it will be created by normal web editors, not just super-techie computer programmers. The web editors will use software (rather like they use page-editing programs such as Dreamweaver today) to create web sites which are capable of providing computers with instructions to do things automatically in conjunction with other sites. The result will be a much greater integration of web sites as agents search through them looking for information relevant to the people who gave them tasks.
This is a different way of thinking about the Web. No longer will web sites be basically stand alone operations (with a few hyperlinks to other sites). They will be automatically interactive with other web sites. But to be interactive they need to be created so that they know what is on the other sites so they can provide instructions to the computers. This will be easily done if the sites are commercially oriented. First, because there will be money to be made in understanding how commercial sites can be semantically linked. Lots of companies will spring up, just like lots of companies appeared when the Web become popular in the mid-1990s. Secondly, the language of commerce is well known. The language of business on the Web has been the object of intense activity in the past few years and can be used to make the connections between sites: the price or cost of this product is this much, in this or some other currency, and the product can be delivered in this way, by this time.
We know much less about the language of labour on the Web. We don't have clear ideas about how we construct our web-based information or discussions. Partly that is because of the invisibility of the LabourWeb. If you can't see the whole of labour's activity on the Web then you can't see how it uses words to describe and present its information needs. The words are vital to the building of web sites which will work on the Semantic Web because they will be the elements which the new coding procedures will use to talk to the computers. For example "contract" will be coded to link to "collective agreement", "expiry date", "recognition clause" and so on. By coding these words according to Semantic Web protocols they will automatically be connected to other sites that perform the same procedure. The key is to know labour's vocabulary.
Fortunately we have a starting point. When SoliComm searches labour sites it creates a list of keywords. As more sites are added to to the searches we will be able to define the vocabulary labour uses on its web sites. And then we can create software tools to enable ordinary union web editors to easily add the Semantic Web codes to their sites. In the end we could have a global database of labour websites which could be intelligently searched by casual users. Union members and the public will begin to see the LabourWeb as a whole and make it work for the labour movement much more interactively and intuitively than today.
A LabourWeb Action Plan
Words are not enough. If the labour movement is to build a focused presence on the Web and prepare for the day it matures into a more intelligently powerful entity we need to start doing what we do best: organize. Now. We need to start organizing the people we have on the Web and the resources at our disposal.
As a first step SoliComm has indexed all the web sites of national federations affiliated to the ITUC and the Global Union Federations In the near future it will include all the larger unions affiliated to those organizations. (Some of which are already included) . The smaller unions will be added to the search path later. This will enable us to search the LabourWeb nationally and globally.
Simultaneously with this process we will need to organize labour activists and subject experts to participate in the educational networks which will run in parallel with the search engine. Then we can work on getting web site editors to categorize their sites. Finally we can start to get unionists to tag the Whole Wide Web.
Is all this possible? Who knows. Is it worth trying? Absolutely.
For more on how people can act as technology organizers see my article on Technology Organizing