-11615

021347

Detours on the Information Highway

Short Stack	We lead with a (very) brief description of the purpose of this note. If you either agree with it completely or find yourself indifferent to it, then skip the rest of the note.
	Purpose	The purpose of this note is to consider a number of the different things about computers in general and the Net in specific that seem to be untrue, exaggerated and/or overblown. There are a dozen or so areas that provoke discussion.
	Argument: I	First, we review some of the comments which were made in a paper somewhat similar to this one written in July of 2002. The conclusion is that most of those remarks have so far stood the test of this short time reasonably well. This is more than can be said about many of the other ideas circulating around that time.
	Argument: II	Next, computers and the net have been touted as being important for education. I think this is a false view, and provide some of the arguments that lead me to believe this. The principal focus of this argument is that most of the arguments about `educational value' are smoke and mirrors, largely pushed by those who have some stake in the process. None of the arguments seem to be supported by any significant research results.
	Argument: III	Third, the importance of computers in `improving' communications has been overestimated. It may have made some things easier and/or quicker, but there isn't much evidence that it's made anything better. While I think I am generally better informed about the `details' of many of my correspontents lives, I think I am less well informed about the broader aspects of their life, so if there is a trade off here it is in favor of `more information about less important things'. This may be a harbinger of other kinds of information in addition to personal correspondence.
	Argument: IV	Fourth, with the price of available storage dropping and more and more software becoming available each day, new products still regularly neglect the problem of making data `exportable' to environments that might ensure being able to take advantage of this long life. This is an indication that most designers apparently are clueless about the real nature of their problems, at least insofar as communciating useful iformation to their `audience' is concerned.
	Argument: V	Next, presentation technologies are supposed to allow us to store information in some generalized structual forms such as XML. This would, presumably, be better. However, most steps are taken in the direction of using storage forms like PDF which actually make things worse, not better. Again, this is a sign of cluelessness about the `real' objective of many of the systems which are being built.
	Argument: VI	Hardware continues to get cheaper and cheaper, but at the same time no more relevant to the central problems of using computers in a more substantial way to process information of broad interest to wider and wider communities. Thus these hardware developments often produce little of ultimate value to those who consume these ever cheaper resources.
	Argument: VII	Software is also being produced at great rates, but most of it seems to be deigned for very naive use, and little of it seems to actually help with any real broad class of problems. And naive use is quite often unimportant or inconsequential use. If particular uses are truly valuable, then we generally do not remain naive for long.
	Argument: VIII	PDAs were a brief vogue. Their time seems to have passed. A very limited version of PDA, the MP3 player, seems to have gained some audience---particularly in the form of the iPod player. However, it is difficult to know, yet, how much of this was due to use of the technology to make copies avoiding copyright. However, PDAs are now much less in evidence than as recently as two years ago, so their acceptance seems to be waining, not waxing.
	Argument: IX	The speed of information flow continues to increase. Yet the usefulness of this, particularly outside of the realm of relatively structured transactions is, at best, problematical. Speed is nice for some professional uses, but often proves not to really matter to the great majority of users. And professional uses have been able to afford the speed---even if it was expensive---for a long time now.
	Argument: X	Audio/Video integration still proves elusive. Much to the detriment of the film based photo/video world, more and more of the images we see are processed principally by computers and passed through networks, but the hardware/software for handling this doesn't seem to be getting much better over time. Professional grade software can be good, but it is generally very expensive, and the limitation on amateur use is generally the amount of effort the amateur is willing to invest in manipulating the images, not the processing capabilities of the systems.
	Argument: XI	The interconnectivity of the net and the fact that computers are becoming very cheap both contribute to suggesting that certain old forms of communication may well not survive far into the twenty first century. Among these are Journals, Conferences and the Academic press.
	Argument: XII	Early adopters are not average people. There is no reason to expect that their attitudes and feelings can usefully be extrapolated to a wider population of the people who follow along later.
	Summary	If you already buy these ideas, then the rest of this paper will probably be boring to you. If you don't care about them, it will probably also be boring. So in either of these cases, skip it. Otherwise, however, you may find something of either controversy or interest in what follows.
Review	I guess it is not fashionable to actually discuss how previous predictions or observations have fared. Perhaps this is because so many of the predictions have proven to be so patently absurd that they make an embarassing example of the expertise of the authors. For example, Doc Searls' suggestion, in early 2000, that This won't be the year the bubble bursts seems rather bizzarely ill-timed. So some quick review is probably in order. This section discusses a few of the things that have been talked about `on the record' either in my web site or on my experimental Antville site.
	Clueless Columns	There are so many `blogs' on the net that no one could read all of them. Worse, the signal / noise ratio is so very low that even if one could read them, the boredom factor would require extreme patience to manage the task. Of the columnists I regularly read, some: Winer and Searls in particular seem to provide a particular picture of cluelessness. Keeping up with and on-going commentary detailing this would be a tedious task, at best, given the loghorrheic nature of their blogs. These blogs have maintained the same low quality of their meanderings that called them to our attention earlier. Nothing has changed here except that the penchant that they show for self-congratulation has become even more obvious as time has passed. On the other hand, it is good that they are self-congratulatory as no one else seems to think much of them.
	Apple Sauce	Apple proved to continue its role as an apparent King of press relations. This seems to be the only way to explain the fact that Apple gets such high marks as an `innovator' in the face of the events that would otherwise suggest that it is certainly a follower, often quite laggard, of the deeds of others. The innovations are such things as creating `Apple Stores'---a marketing strategy that failed more than a decade ago---adoption of operating systems which were developed elsewhere and which are ancient. Also Apple continued a rather bizzare advertising campaign, based seemingly on the premise that ignorance is bliss. The stores, while they continue to garner glowing press reports, seem empty every time I visit one, and their indicated `gross sales' at under $3m / store / yr would appear to at best barely cover costs, even before their negative impact on the sales of surrounding non-company owned sites is taken into account.
	The Cluetrain Stops Here	A new copy of the Cluetrain Manifesto now lists at $2.45, and is overpriced at that level. This book enjoyed a blessedly brief vogue, and has rapidly slipped into the oblivion that it so richly deserves. An open invitation to all kinds of jokes about `cluelessness', this manifesto advances 95 theses (of the intellectual caliber of We are immune to advertising. Just forget it.). It reminds me of the motto of the `Tortola Golden Striders' `Starts slowly, then eases off.'
	Sales Collapse	The sales of hardware, which supposedly were going to begin to recover in 2001, failed to do so. Then, 2002 became the designated year for the `recovery'. But that passed without any improvement. While we now hear that this will happen real soon now this faiulre isn't exactly a surprise in a market place where few users consume any significant fraction of the resources that they already have at their dosposal. In addition, as prices continue to fall, unless demand increases faster than the rate at which prices fall, revenues are bound to suffer. Companies like Apple are facing truly lackluster sales, and now make substantially more money on the investment of their cash portfolios than they do making either computers or software. In the long run this does not make them particularly attractive as investment candidates. Everyone seems to still be waiting, increasingly anxiously, for the killer application, but it so far has failed to show up even though there is substantial use of computers in some new problem domains: photos, music and video just to mention a few. What these tend to prove is that most people are neither good photographers nor good directors, and better technology doesn't deal with the problem at the root: if you don't have anything to `say', better technology won't help you much.
	Summary	Not everything predicted came to pass during the last six months, but the record is pretty good. There don't seem to have been many provable misses, and although all of the issues are not yet resolved, the ones we have had a chance to evaluate have pretty much proven true.
Education	I find an eerie parallel between the much touted use of computers in education today with the expectations about television that were quite common in the 1940s and 1950s. I attended a High School that was the site of a good deal of research on the use of TV in education. Up thru the 1950s many people believed that television was going to revolutionize education, and confident predictions were made that a major part of the educational program would ultimately be handed over to TV, to be managed remotely. For the most part, this never happened. Indeed there is reason to believe that TV, instead of providing a tool for educational progress, has instead made a negative contribution both by influencing expectations for speedy reward and by replacing educational values with entertainment. There is increasing suspicion that the same thing might be happening with respect to the Net.
	Education and Computers	It has been widely assumed that computers are an important element of an education. While there seems to be remarkably little evidence to support this fact, at least in the form of documented research, it nevertheless has become a widely shared tenet of education. The notion, I guess, is that computers are somehow a new form of research tool. It is also perhaps relevant that most of the current generation of teachers did not have a great deal of access to computers during their education, and this makes it easy to overestimate their value when applied to education. It is generally easy to be optimistic about the likely impact of future technologies, particularly when we are ignorant of the facts.
	Education and The Net	The net adds an important dimension to the potential use of computers in education. If the computers provide the underlying technology, the net provides the essential fabric of information that seems to have such an important educational component. However, a lot depends on just what part of the `educational task' we are talking about. We need to carefully distinguish between the part of education that deals with `fact access and retrieval' from that which focusses on how we process the facts once we have them. While there may be some argument that the net helps us with access to the facts, I find absolutely no evidence that the net has had any favorable impact whatsoever on our ability to process the facts which we acquire.
	Summary	Many people believe computers are a necessary part of `modern education'. I do not. Worse, I find that there is little evidence that they help, and there is considerable evidence---albeit in unorganized form---that they may actually hurt. The can be distractive, and they all too often replace fruitful and productive arguments about the substance of the matters that are the subject of the education with (largely) pointless and very short lived considerations of the technology used to access the facts.
Communication	It seems clear that computers have had a substantial impact on communications. The most often cited use of the net is for `keeping in touch' using EMail, or---if you are talking to grade school and high school kids, through various `Chat Groups'. A great number of people no longer write letters at all: they simply write brief EMails to keep in touch with their family members and friends that are scattered all around the world. This truly is a new capability, not present for most people as recently as half-a-dozen years ago.
	Personal	There are at least three different kinds of personal communication which are supported by computers and the net. They differ principally in the nature of the interaction that they define.
		Blogs: Blogs (Weblogs are as `individual' as the people who write them. This means that they range all over the map, some quite good and many that are terrible---probably not far from the ratio of good/bad that can be observed in many other fields of endavour. If we look at the paintings done in a local YMCA class or at a local high school, we will find only very few that are of much interest to other than the close family members of the painters. And many don't even pass that test. Similarly for the piano pieces played at a local music teacher's recital or in the writings that might appear in a small local newspaper. So it's probably not so surprising that most blogs are bad. These days, if you shop around, you can probably find some place to house your blog for free, and at worst, by spending $25 to $40 per year you can be `published'.
		EMail: EMail is another form of communication that is very popular. Since many college kids these days either own their own computer or have free access to computers on their campus, a great portion of the communications with `home' that aren't handled by long-distance cell phone are written and sent via EMail. It is not uncommon for there to be almost daily, if brief, contact between parents and their college-age children, something that is a far cry from the once-a-week phone call that might have been more common during the middle of the last century when I went to college. The fact that EMail is increasingly ubiquitous does not, however, make any comment on the quality of the information which is exchanged. In my experience, EMails tend to be a great deal shorter and a good deal less `philosophical' than my letters used to be. EMail is often dashed off in a brief free moment that happens to occur between two otherwise pressing activities. They are often brief, factual and not particularly thoughtful. It proves to be a good medium for exchanging an important class of straightforward factual information, but not for expressing more profound reflection. An equally critical aspect of EMail is that it is `store and forward'. It can be sent when the sender finds convenient, and is typically read only when the reader finds it convenient to read the EMail, often in a whole batch of arriving messages. The time delays between exchanges are thus irregular and sometimes frustrating.
		Chat: Chat is rather different kind of communication. Chat only takes place contemporaneously, between two (or more) people who willingly share a particular time. For two way communication, it would appear to be a rather inferior form to a conventional phone call, from which it differs in only a few ways. The differences are that the parties may more easily pay attention to an alternative agenda when chatting by computer than when they are both on the phone. In addition, those chatting do not expose as much `of themselves' over a terminal as they do by phone. Thus it is easier to `spoof' during conversations, and for whatever reason, people seem to be more willing to engage in conversations with previously unknown parties by computer than they ever would be willing to do by phone.
		Bulletin Boards: Bulletin boards are still another form of communication. They are, in one sense at least, `EMail to the world'. In other words they are a form of EMail where the message is not directed to a particular addressee, but rather to a board where it may be read by others who regularly pass over the content of the board looking for things of interest. Bulletin boards---or their close equivalent maillists---seem to work best in communities that share some common focussed interest. They prove to be quite an effective way of exchanging information about a particular programming language, for example. The disadvantage of bulletin boards tends to be the fact that sometimes they attract destructive individuals who have more interest in `trolling' or provoking controversy than in exhanging useful information.
	Journalism	A lot of attention has been paid to the role of the web in exchanging information that we might previously have expected to exchange in newspapers or on television.
		Diaries: Diary bogs are of two general types. The general blog covers most of the aspects of the life of the blogger, and is a lot like what we think of as a conventional diary. Other blogs focus on some particular area that happens to be of particular interest to the blogger. The special interest blogs are, perhaps not surprisingly, often about technological matters. This isn't much of a surprise given the fact that early adopters of blogging technology are quite likely to be interested in other aspects of that technology as well. I won't address the issue of the quality of the general blogs. Whether you like them or not is very much a matter of personal taste. To my taste, most of them are somewhere between terrible and mildly interesting, but you might find them much more to your taste. The technologically inclined blogs, on the other hand, generally purport to discuss topics of some technological interest. Very occasionally they do. However, I find most of them to be of dubious value---often quite ill-informed.
		Live Blogs: For what now appears to have been a very brief while, `live' blogs were a focus of some particular interest. This idea is straightforward enough. Modern technology---including wireless capabilities and highly portable computers---make it an easy matter to write entries to a blog while a panel discussion or speech is actually in progress. Since modern laptops are quite silent, and since interconnection to the web is now quite easy, this is something that it is quite easy to do. And for a while it was done quite a bit. However, it rapidly became clear (this is my opinion, at least) that most of the information presented in the typical speech or panel was not of such a desperately timely nature that much was accomplished by circulating the information immediately. Indeed, most of the panels that were blogged were incredibly dull. As a result this seems to be dying out. But who knows, it will take some time before we can be sure.
		War Blogs: `War' blogs are sort of a special case. But they are numerous enough to suggest some special mention. This sub-category of blogs spends most of its `ink' on political subjects, particularly focused since 9/11 on `war'. They are both pro- and con- the possibility of `war' as an instrument of foreign policy, and are particularly characterized by a general lack of temperance in expression. This appears to be a fairly common characteristic of some of the communications on the net, namely that they are not much constrained by the civility that seems to bound and govern more personal forms of interaction. There are some `mixed' blogs, of course, where the authors will spend most of their time writing about their lives, but then will occasionally lapse, when the spirit moves them into what often proves to be a logarrhyic tirade on some aspects of world politics.
	Corporate Nets	Corporate nets are more and more common everyday, but are not much discussed in the public provenance. I guess this is no surprise. Over the past decade, as the public has made wider and wider use of the net, corporations have followed along in the realization that this might be a very useful technology for communicating between disparate parts of the corporation. The principal problem involved in doing so, of course, is security. Most of the messages that most of us pass on the net are of low consequence and value. While we would occasionally be embarassed to have these communications made public, usually there would be little result other than a small amount of embarassment associated with this occuring. And while there is occasionally some monetary `value' associated with these communications, it usually is small enough that it would be unlikely for it to be a target of any organized effort to discover their content. For a company this is not so. Information passed on the net might have consequential value, and might very well be the target of some form of industrial espionage or outright theft.
	Security Issues	The `security' issue for the Net appears to divide itself into two major parts: corporate and personal. The corporate aspect of security is often handled through VPNs (Virtual Private Networks) or similar technology. This involves using perfectly conventional net messaging but it has the added twist that the net mwssages are encrypted in such a way that if they are intercepted and copied during their transmission, nothing useful will be discovered. It is pretty much just as though you were working on a private network. The technology for handling personal communications with encryption adequate enough to keep messages private has been available for decades, but it has only found a very limited use. While it can be done, few people find it worthwhile to actually bother to use the technology, and those who do generally quickly earn a reputation as a `crank'.
	The Demise of AOL	The announcement of Steve Case's forthcoming resignation from the chairmanship of AOL represents a public admission of something that has been fairly obvious for at least a year now. Namely, that the notion that there was value in controlling the access portals to the net was vastly overvalued. At the time of the merger to AOL and Time-Warner the corporations had market values of approximately $200b and $120b. Now the market values the combined operation at about $67b. The collapse in value is indicative of more than just the mis-management and over-sell which was very much a central part of the AOL / Time milieu. It represents a sharpening of the valuation of the decrease in overall value in the control of information channels that is very much a part of the fact that the net now makes moving information around, and dispersing it widely and quickly, a much cheaper operation than it used to be. This particular story is relevant here because it suggests the fact that the value of the control of the flow of information was wildly overestimated during the dot com boom. The assumption was that somehow control of the access portal would somehow allow value to be extracted from accessing the information that was assembled for more convetional journalistic purposes. As the collapse of value indicates this did not prove to be the case.
	Summary	Modern networks allow us to both access information faster and to move it around both faster and much more cheaply. However, this doesn't necessarily lead to any improvement in function. When we can only move information slowly it is easy to fall prey to the assumption that if we were able to speed things up, we'd be better off. However, often this is not the case. The limitations that we face only occasionally have to do with the speed with which we can move information around, and in all the cases that don't, speeding up information flows won't have any benificial effect.
Pornography	Realistically, it would appear that the one unqualified success of the net is, for better or---more likely---for worse, pornography. This fact is only rarely discussed.
	Profits and Pornography	Not much on the net has actually proven to be profitable. The conventional media, for example, have been trying, with at best limited success, to sell access to their information on the net in a way that will not cause conflict with their other more conventional business. A considerable number of highly touted publications have attempted to make a go of this kind of business, and to this point in time, success has proven to be quite elusive. The exception to this, apparently, is pornography---which seems to be a rather unqualified success on the net. It would be interesting to engage in speculation about whether there is more to the net's success with pornography than a simple issue of anonymity, but that will have to await some psycological studies that will probably be made over the next decades. For our purposes here it is only useful to note that this is one place where we have an effective and dynamic marketplace established very quickly by the Net, and as of now it shows no signs of easing off.
	Summary	Whether there are more marketplaces that will prove to have the dynamic character of this one, only time will tell.
Import and Export	Network sofware seems to be a little casual with respect to managing import and export of information. If we desire our information resources to be available across a substantial period of time then it is necessary that we pay attention to how we manage to move information in to and out of the software that we use.
	Long Life	One of the most important properties of information that we might want to access on the net is that we are able to archive and access it over long periods of time. Yet most of the software that is produced to help us manage this information seems to use idiosyncratic file structures and, worse, has no import/export facilities that allow us to use other systems to help us manage the information. This is a critical error. One would have to be crazy to commit any important information to a system which cannot guarantee a long life. Information which is stored in idiosyncratic file structures requires that not only the software continue to be supported, but it also demands that there is hardware available that can run this software. Over the horizsion of a few years this probably isn't a problem, but over a longer time period it becomes quite problematical. Think, for example, of all of the information that was committed to files supported by software such as Electric Pencil or Edix/Wordix. These were popular pieces of software in the early and middle 1970s. Not only have the companies that built and maintained this software long departed from the computational scene, the programs all ran on CPM, the `Windows' of its day, which no longer runs anywhere. Probably one could, if were absolutely vital, take a few weeks or months to find and restructure such a system, but at best it would be a very expensive proposition, and at worst it might well prove effectively impossible, as the media used to store the information is quite likely to have deteriorated sufficiently to no longer be readable.
	Importance	Another point which deserves to be made is that it would be foolish to commit any important data to storage forms that are both out of one's own control and which have only limited usefulness. The more important the data is the more important that we maintain control over it. I cannot imagine comitting any serious effort to information which will have a usefulness which is limited by the current conception of some systems designer that has no specific understanding of my problems.
	Lock-In	Another thing which is relevant in many circumstances is the `lock-in' which is caused by maintaining important data in a form which is under someone else's control. No matter how good and how helpful a piece of software is today, there is always the possibility that tomorrow someone else will come out with something which is even more useful. If you are locked in to a particular piece of current software with some important data, then you will not be able to take advantage of any new developments without incurring substantial costs.
	Control of Information Flow	Another things which is worth paying attention to is the fact that the net makes it more and more difficult for us, or anyone else to maintain control over the flow of information. Information can now move virtually everywhere with great ease.
	Summary	Import and export of information in an easy to read and easy to manage format is vitally important, particularly for information which is of critical interest.
Politics	Politics represent a `new use' of the technology of blogs. Probably because it is new, some have jumped to the immediate conclusion that it is also effective. This is not so clear.
	Tara Sue, Boo Hoo Hoo Hoo	During the 2002 elections, the first `political' blogs appeared on the scene. It became fashionable, in blogging circles to make a big deal out of the battle in one particular North Carolina district where a veteran Congressman was running against one of the supposed `new wave' of candidates with a blog. Much was made of how this was to be the `wave of the future'. Unfortunately, however, the results just didn't pan out. The candidate in question recieved less than half as many votes as the former Ohio Congressmn who, as a convicted felon, was actually running from his jail cell. All-in-all not a high water mark for the effectiveness of blogs. It could just be that Blogs, rather than being an effective vehicle for reaching some of the voters, are actually a waste of time, consuming more time and energy with less effect than other more conventional means of campaigning. All of this is still new enough that it probably would be wise not to draw any unwarranted and unsupported conclusions, but it does seem quite safe to say that it is not obvious that blogs are a good tool for practicing politics.
	Summary	It will take some time to understand the real possible impact of blogging on politics. Wisdom would suggest not leaping to conclusions too quickly.
Presentation	One characteristic of the net is that there are many different systems which connect to it. Each system has one or more pieces of `browsing' technology which are used to access and view the information. This creates a considerable presentation problem, and there are a number of different schemes which have been used in an attempt to deal with this complexity.
	HTML	The invention of HTML represents a particularly important step in this process. HTML is rendered with reasonable consistency across a wide range of systems. It is a simple and effective language that handles straightforward display problems quite well.
	XML	By this time, many people assumed that XML would have assumed a leadership role in the structuring, archiving and transmission of information. While it seems to have an increasing role, XML is still only scantily used, particularly in contrast with other alternatives. As with most things in the computational milieu, there are both advantages and disadvantages. The advantages of storing some structural information along with the data are obvious. We are able to recognize the significance of the content of blocks of information in a much wider context than just where it was originally planned. However, this brings along some concamitant disadvantages, the principal one of which has to do with the overhead implicit in storing some representation of this structure.
		XML vs. HTML: There is a trade-off between XML and HTML. HTML only deals with the display of the information, there isn't any reference to the structure of the information that is being displayed. XML adds this second dimension to the display process. In a sense the difference is between a one level display process and a two level display process. The first (HTML) simply describes how we want to see the information. The second (XML) allows us to describe the structure, and how we want the structure to be displayed. Then we describe the data in its structure, and this allows the re-mapping to take place that actually generates the display.
	PDF	PDF represents another alternative for the long run storage of information. This particular format is touted (particularly by its proprietary inventors) as a wonderful innovation. A little inspection, however, suggests that like other things it is really a two-edged proposition, trading off the advantages with some distinct, and---in my opinion at least---more significant disadvantages. The advantages generally have to do with document portability. Files in PDF are widely viewable in a number of different computational worlds. This is accomplished, for the most part, by sort of making the document into a `picture' which can then be rendered, like most pictures, without any reference to other information. However, the fact that a document is a picture also means that it is more difficult, if not impossible, to access the `information' which is contained within the document for any purpose other than simple viewing. To understand this, think of a file representing a train schedule. If we have such a document as a `picture' we clearly can read anything in the schedule by simply using our eyes. However, if we want to answer some question which is not directly presented in the document---for example the duration of a particular trip, then we are in trouble. We have to use our eyes to extract the information before we can perform the appropriate calculations. If the information were there as information, then we could probably figure out how to do this through a computational process.
	Scripting Languages	A whole range of languages that allow the dynamic interaction with browsers have been invented and developed. Perhaps the most important of these is Java, but PHP, subsets of perl, Python and other languages are examples. These languages allow interaction with web pages that ranges from simple form filling to highly complex data base access and display.
	Summary	The struggle between various ways of reprsenting information on the net is just getting interesting. If the number of different ways to solve this problem continues to proliferate, that may not augur well for future prospects, as supporting a large number of different formats will become an increasing burden on software resources.
Problems	In addition to some of the things we have been discussing, there are also some idiosyncratic problems with the net. As time passes some of these things will be an increasing burden on the operation of the net.
	Dead Links	Dead links are becoming an increasing problem. When all web pages were new, all of the references were equally new, and therefore all of the references `worked'. As time passed, the number of pages grew dramatically, and pages begin to `die'. However so long as the volume of new pages was growing at a high rate the proportion of dead pages remained quite low. However, as we pass the inflection point in the growth rate, the rate at which pages die will begin to cause the proportion of dead pages to begin to rise. Since it is frustrating to access a page only to find that the page is not there, this will impose and increasing burden on the effectiveness of the system.
	Summary	The honeymoon period is over. As time passes problems will be an increasing part of the future. Ultimately it will things like these that maylimit the effectiveness of the Net.
Hardware	Hardware prices continue to fall, but not in any way that provokes much that is interesting. Storage, which looked for a while like it would become a problem item because of the rising demand for space due to photo, video and music storage, seem to have fallen in price fast enough to keep from being a problem of any particular substance. Storage costs have dropped down to nearly $1 / 1Gb, and this means that a CD can be stored for about $0.50 and a movie can be stored for only $5 or so.
	Size Limitations	We seem to be getting to the point where size is limited by human form factor considerations rather than any technology. Of course, there is always the chance that there will be a technological breakthrough that happens to produce an effective solution to some of the still outstanding problems, but that is not something that we are likely to be wise counting on. We can now build devices that carry complete and flexible operating systems that are the size of a matchbox, and that weigh a few ounces, so size will tend to be dictated by human convenience rather than electronic conisderations. Important factors which impact size are Size of CD/DVD; Size of Keyboard; Size of display; Size of other devices and connections. Quite obviously there is no `right' size---each person / situation may find different tradeoffs effective.
	Apple	Apple's much touted design for the iMac proved, as I had suggested in early January, to be a disaster. Unsold iMace are clogging the shelves of the Apple distributors, and sales of everything---with the possible exception of iPods---have been flat or worse.
	OQO Computer	The OQO was supposed to be a harbinger of a new breed of PDAs. This paperback book sized device supposedly will have a `real' Windows operating system and a multi-gigabyte hard disk. The size of the screen is dictated by the form factor, and the input is captured by marking the screen. A conventional keyboard and display is supposed to be available as a briefcase-like add-on. The device is also going to have connections for most of the conventional communications: Ethernet and Bluetooth. It is targeted to weigh about 9 ounces. That's the good news. The bad news is that the device has already slipped by at least one quarter, and if the web site is any indication, there not only is no discussion of this fact, but there has been no detectable change for many months. Thus the arrival of the device must be suggested to be treated as problematical.
	Sony's Desktop	There is one new development in hardware. Sony has released the first device which, in my opinion at least, which emphasises the physical look of the machine to the extent that it becomes a piece of furniture capable of being displayed in a living space. This is a departure that dramatically surpasses even Apple, which has always had an emphasis on the physical design of its hardware. This box looks like a futuristic clock when the keyboard is closed. By folding down the keyboard, the screen lights and the computer function takes over. This would have been an even more optimistic development if Sony had chosen to provide some capability which would link the device into the audio / video world of the kind of high fidelity systems that Sony produces in its home electronics division.
	Wireless	We appear to be just at the cusp of wireless becoming an important mode of local area net interaction. While direct internet access via wireless is still generally restricted to fairly low bandwidth, local area network wireless rates have risen from 11mb/sec. to 55mb/sec. thus affording new capabilities and options.
		802.11g: It would appear that 802.11g will soon be widely available. This is particularly important as we expect to transition from 802.11b (11mb/sec.) up to the higher speeds. While 802.11g and 802.11a both have the same maximum speed, 802.11g has a compatability advantage in that is also compatible with 802.11b, and thus makes it easier to effect the partial upgrading of a system.
		Bluetooth: Bluetooth networking is fairly new, and for the most part it currently deals with linkages between computers and things like phones and a few of the newer printers. The advantages / disadvantages of Bluetooth vs. are not yet clear, but some common practice should begin to emerge in 2003.
		Warchalking: This is a current vogue for finding places that allow free access to the internet via a wireless connection. While gathering this data may be of some interest, and can represent a game not unlike GPS-caching, my guess is that it doesn't represent any sustainable value. Either access is of some value to the user, in which case companies are likely to begin to charge for it, or it isn't of much value, in which case it doesn't matter much whether it is charged for or free---the there isn't much to choose from either way. I would expect Warchalking to have only a brief life of interest.
	Mass Storage	The picture of mass storage has changed a great deal in the past few years. Over the relatively short period of time represented by a couple of years, large storage has dropped from about $1000 / Gigabyte down to $1 / Gigabyte. This provides some dramatic opportunities for storing and manipulating huge quantities of information. Storing a whole CD now costs about $0.50 and storing a whole movie is a matter of a few dollars. When Microsoft first made its terabyte map server public the storage probably cost tens of thousands of dollars. 18 months ago a terabyte server could be built for $5000 or so. Today we could probably put one together for something more like $1000. When storage costs change this dramatically, and when we have very cheap `pipes' (The Net) that can be used to fill them up, all sorts of new opportunities arise that might be worth investigating.
		Portable Storage: Portable storage has also dropped dramatically in cost. Portable MP3 players, for example, rapidly moved past the 5 gigabyte capacity to 10 and then 20 gigabytes. It is reasonable to expect them to continue further in the not very distand future.
	Summary	Hardware continues to develop, and more capability is delivered at lower cost everyday. However, in recent times none of these developments seem to be intrinsically very interesting. Machines keep getting faster, and storage keeps getting cheaper. There are certainly innovations at the margin, wireless being one of them. But cumulatively these seem to have startlingly little impact on what is being done and how it is being accomplished.
Software	If hardware innovations have been of only limited interest, very much the same can be said for software.
	Scripting Pages	The management of web pages has become an enterprise in itself. There are several scripting environments that let users develop and manage their own web pages. For the most part these systems are technologically unsophisticated and intellectually uninteresting, although they do deliver the desired capabilities with straightforward effectiveness.
	Hosting	The delivery of web pages is really a two stage process. First, the pages have to be designed, created and organized. Second, the pages that have been developed have to be housed and given URL addresses on the net. Sometimes these two capabilities are delivered as a part of the same package, and sometimes they are quite separate.
	Data Base Management	While data base management technology is in wide use, some of it applies directly to the problem of producing web pages. Lots of web pages contain regular segments of information which are best managed as `data'. For example, each of the entries in a diary log are quite likely to `look' alot like one another. Rather than restructuring a page de novo each time we want to add and entry it is easier if we can add the entries to a data base and then generate the appropriate page displays. Many of the scripting products that allow for the building and managing of web pages provide an interface to just this kind of facility.
	PIMs	There continues to be a struggle to produce a Personal Information Manager that will catch on and have a life beyond a few years. Lotus tried with Agenda, and failed. Ecco was successful for a while, and then it failed. We are regularly treated to software that is supposed to deal with this problem, but so far none of it seems to live long enough to really provide any durable interest. My speculation is that the technology, for all it's apparent `showy' value, actually has only a few advantages and, at the same time, an overwhelming number of disadvantages when compared with the simplest `pencil and paper' systems, and that this may well account for the lack of any long run acceptance.
	Outliners	Outliners are another case where we may be running into a `more is less' phenomenon. Bloggers, in particular, seem unduly attracted to the wonders of the outline. Yet there is no indication whatever in my experience that the documents produced from outlines are in any way superior to those which are not. Indeed, if the output of the bloggers who spend the most time talking about outlines is any indication, the results are distinctly not stunning: or in other words, nothing to `write home about'. Outliners are an `organizational' tool, and thus it wouldn't be at all surprising if they particularly appealed to the compulsive people who are, perhaps, more likely than average to be interested in blogging.
	Summary	As was the case with hardware, there's very little of interest that is appearing along the software front. Developers keep taking stabs at some of the problems, but they seem to be, for the most part, simple retries of solutions that have been attempted, and failed, before. It is perhaps revealing that it was regarded as news when Mitch Kapor, the original developer of Lotus 1-2-3 and later the Agenda PIM has announced that he is going to try again with what appears to be very much an Agenda retry.
PDAs	PDAs were once popular, but may have passed their high-water-mark of acceptance, at least for a while.
	MP3 Players	One of the successes over the past year or so has been the development of the MP3 player. This can be viewed as a specialized form of portable device which is designed to allow access to a common form of data file which is used to contain music. For a brief period of time it was a vogue to exchange files of music. This effectively worked around copyright laws and led to the wide availability of very cheap music. Whether the widespread use of MP3 devices is in any way dependent on this relationship to the copyright situation.
	The Cell Phone	The relationship between the cell phone and the PDA is also undergoing an evolution at this point in time. Cell phones are getting more and more sophisticated while PDAs are beginning to develop wireless capabilities. Indeed a few devices that exactly span this problem domain have been announced.
	The Downside	Electronic PDAs have a downside. So do paper documents. But the ratio is uneven. In my lifetime I have lost one appointment book over about 40 years of an active career. I have lost the contents of various PDAs at least a dozen times in the past decade. So the score has been quite uneven.
	Summary	PDAs took about two years to go from nowhere to somewhere. Then they took about another two years to go from somewhere back, pretty much, to nowhere. For a brief while it was common for everyone in a business meeting to be sporting the latest form of some PDA. People with their Fil-o-fax seemed to be positively retrograde. Then this began to switch back. Paper schedule books began to reappear.
Speed of Information Flow	Information flows around the net with great speed. Many net users are not really used to this yet. As a result it is easy to make mis-judgements about the nature and impact of many of the phenomena that are encountered in that domain.
	Markets can Vaporize	One characteristic of the Net which hasn't been factored adequately into lots of the analyses that deal with net-related things is the fact that net marketplaces can vaporize with a rapidity that is stunning. If you think about it this should not come as a surprise. One of the most important underlying characteristics of the net is the rapidity with which information can be moved around. But this isn't a one-way street. Markets can be created overnight, but they can equally well disappear overnight, as well. And this fact has some economic implications that need to be throught through.
	Volatility of Valuation	The fact that markets can evaporate makes valuation incredibly volatile. In past times, for example, the `value' of a banking institution was in no small part dependent on its physical presence. A good `solid' bank building implied good `solid' resources backing up the bank.
	Scanning for Prices	The ability to scan the net for prices provides one example of something that contributes to the volatility of the net. In the `old days' one could support different prices for something in different parts of the world. Indeed, it was often possible to support different prices in places as close as other parts of town. And if prices changed, one could reasonably expect to be have some time, often days and sometimes weeks, before the new prices would be `discovered' in places that were far removed. This is no longer the case. A price change may be noticed within minutes of making it public.
	Summary	The speed with which information flows around the net changes lots of economic considerations dramatically. It will take a while to sort out the long run effects of these changes, but there is already some indication that we will need to learn to live with a dramatically new set of economics.
Self Aggrandizement	One of the most trying aspects of the web, and of blogs in particular, is the tendency to self-aggrandizement.
	Summary	I guess it isn't a surprise if people who are interested in `publicity' would themselves seek publicity. Only time will tell if this tedious focus will ebb.
Audio / Video Integration	In the earliest days of the development of small computers there was a direct relationship between computers and television. In the earliest days it was commonplace fo computers to mix their display output so that it could be `broadcast' to a TV set (by cable). While this was cheap, it also did not allow sufficient control of the images, so eventually a decoded signal was passed, allowing more precise control. Similarly for sound.
	CD/DVD	CDs and DVDs have become a common storage form, not only for music and images, but also for information on computers. Laptop computers with DVD players are a fair trade for the DVD players on that are a part of a conventional TV.
	TV	As LCD/Plasma TVs have come into existence, there has been some return to merging the role of the TV screen and the computer monitor. As computer prices have fallen, there has been some shift to begin to use LCD/Plasma screens instead of monitors.
	Photos and MP3s	Digital photography and music have also pretty much replaced the analogue measures
	Control	Modern hifi systems are becoming bewilderingly complex. A typical setup can have half a dozen or more remotes, all of which interact in some complex way. The manuals can easily mount to hundreds of pages. Using a computer control some of this offers the prospect for re-phrasing the problem and re-casting it into one of a different kind than can easily be handled by buttons on a remote.
	Copying and Copyright	But the fact that images and sounds are now all in a digital form raises a whole dimension of complexity for copyright issues.
	Summary	The integration of computers and audio/video processes is taking place. It isn't yet clear just how far this will go---it may ultimately merge them all into the same milieu.
Walking Corpses	While there are some things that the net makes possible, there are also many things that it effectively destroys. The communications channel created by the net affords a wide opportunity to perform many selective, broad, communications tasks at very low cost and with considerable ease. Because it does this, it also removes the economic justification for several time honored communications tools.
	Journals	One example is expensive journals. Prior to the existence of he net they served a vital function. Acadenuc information was exchanged in journals. They had a circulation which was broad enough to reach academics doing research in related areas. Because the production of the journals cost a lot per user however, they were almost invariably expensive. The net affords the opportunity to accomplish a similar distribution of information with two very important differences: they are both much cheaper and can collect, compose and distribute the information much more quickly. We can now circulate our research results around the world within minutes of preparing them. And the cost of doing so may be essentially zero give current cost structures.
	Conferences	Conferences may also be of declining importance because of the opportunities afforded by the net. On line meetings, and the within the day (it can even be within the hour) circulation of documents has become a commonplace. The expense and danger associated with current day travel makes on-line conferencing and exchange of documents a very practical possibility.
	Academic Press	The academic press is another place where we expect a substantial decline in a marketplace as academic publications are replaced by the distribution of information on the net. Academic publications are expensive, and quite often not particularly well-produced. Distribution costs are substantial and publication delays can interfere with distribution to a very material degree. Publication on the net seems to solve all of these problems. It is both low cost and fast. It also proves to be very effective.
	Summary	So the web has created some walking corpses. It would be foolish to try to predict with any accuracy when it will be time to bury them, but eventually they will wither and waste away.
Early Adopters are Different	Many of the assumptions that circulate about the likely future growth of the net are based on a false premise. This premise is that we can get some notion about the likely behavior of the future prospects for net functions and net business opportunities by extrapolation from current use. Because of the very technological nature of the net this is a dubious proposition at the very best. There is little reason to believe that there is anything `typical' about the early adopters of this technology. And useful extrapolation from one population into another can only be made if the populations involved share common elements. If the populations are quite different in significant dimensions then little can be said. This is the case with the net. Early adopters are a very unusual crew, and are not very likely to use the technology in a way that people who follow on later in the cycle are likely to use the technolgy. Early adopters are often better informed, more curious and less destructive than many of those who will follow them.
	Hawthorne Effect	One of the dangers of extrapolation from early use involves the Hawthorne Effect. This phenomenon, first encountered at a plant in Hawthorne Massachusetts, suggests, among other things that in the early stages of evolution, people are likely to be more focussed and interested than may well prove to be the case in the longer run.
	Examples	Examples of this problem occur in many different contexts. Sometimes they are due to a self-selection of an unusual population of early adopters. Other times they are a simple result of growth in the user community. Things which work in the `small' do not necessarily work in the `large'.
		E-Mail: E-Mail provides a good example of a situation where the early adopters may be a different population than those who follow on. When we started using E-Mail regularly (this was in 1973) all E-Mail messages were interesting. After all, they were all written by people who were working on research projects similar to ours. So naturally what they had to say was of considerable interest. As time passed, however, and the user community broadened, E-Mail became much more like all the rest of the mail that we processed. It was no more or no less interesting, just a different form.
		Course Administration: Another example illustrates the growth phenomena. A few years ago, a few courses at many universities started to use computers to help in the administration of course materials. This was both novel and effective. And the novelty provided reward enough for the trouble involved in learning new ways to deal with these problems. As time passed, however, course after course was added to the list, and---since they were typically developed quite independently---each introduced its own way of dealing with these matters---so this began to become a distractive burden.
		Blogs: Blogs are another example of both of these notions. First, the original bloggers were an `unusual' lot. The they shared many common elements of interest: Technological Adequacy; Interest in self publicity; Enough free time Second, at the outset there weren't many bloggers, so one could reasonably look at a fair percentage of them and choose to follow the ones that were of some interest. A third factor, which won't be analyzed in depth here, is the arrival of a `story' (in this case the Clinton Impeachment) that was almost ideally designed to focus interest on the method of information exchange because of the wide interest in the subject matter of the various investigations and disclosures.
	Summary	Early adopters are `different'. Generalization based on observations of their behavior are dubious at best.
Conclusion