September 2009


Sep 13, 2009: 24. Musings on computing and a new internet


Since my health remains bad, this section is not often updated or extended, but here are a number of new things relating to computing that I recently found and liked, and an urgent proposal of another internet.

2. Firefox
         AddBlock Plus
         Nuke Anything Enhanced
3. JavaScript
          JavaScript Debugger
4. Smalltalk news
5. Open source
6. A new internet is needed + 3 special reasons why
           First - it is and ought to be public and open
           Second - it makes solid sense
           Third - it may save the public media

1. ADSL:

Having a bad health since over 30 years (I have ME, which also means that typically you won't get help at all, especially in such enlightened countries like the US and Holland), I am also poor, so it took some time to upgrade from a telephone-modem (yes, I worked with that until the summer of 2009....), but I did it, it works, and it is a vast improvement - with a considerable problem.

The improvement in speed and ease is obvious, and pretty spectacular if you were used to a 56 Kb telephone-modem (that often worked at 4 Kb speed, because the Dutch provider xs4all seems to like it that way, or at least doesn't care shit, and doesn't answer complaints and queries of knowledgeable folks: I have reams of unanswered mails, and that's why my site is since 2004 maintained at one.com, which is orders of magnitude better, and perfectly decent in dealing with complaints - and I have had no serious problems with them over the last five years at all: They simply work, and do it well, whereas xs4all doesn't work, or if they do make a mess of it (*)).

The considerable problem is not only McAfee that comes with xs4all, (and that tells me my computer is all hunky dory, fine and healthy whereas my own experience and other software tells me it is not) but in general that (1) with ADSL (and better) you loose part of your control over your own computer when connected to the net, since everything happens very fast, and mostly undocumented, and (2) there is no decent doable way for private folks to do much about it, even if they know a lot about computers and computing.

I guess this is more of a Windows than a Mac or Linux problem, but I use Windows, mostly for practical or financial reasons.

Anyway... ADSL is nice (and lots cheaper than a telephone-modem, it turns out), but it comes with dangers about which one cannot do much oneself, at least not on Windows.

But there are some things that help, and to these I turn now - and I trust that if you are sufficiently interested in something I mention, you know how to find it on the internet, without me having to supply links.

2. Firefox:

Firefox - the most recent version is 3.5.3 simply is a much better browser than MS Explorer. It is better organized, easier and more pleasant to work with, and also it has numerous fine extensions and Add-Ons AND it is both free and open source.

If you don't use Firefox, my advice is you should: Installing is a breeze (with ADSL, to be sure, but then it's done in a few minutes), and once you have it it is wise to turn fast to the many (nearly all free, open source) Add-Ons that come with it.

There are very many, for many ends and purposes, and I will here list just three that contributed much to my piece of mind when browsing the internet with Firefox: AdBlock Plus, Nuke Anything Enhanced and Noscript.

Adblock Plus gives a considerable amount of tools to block the downloading of ads in your browser. It does this fairly well and effectively, and the only setback I have seen is that to use it sensibly you need to have some basic knowledge about html. Enters the next Add On, that I find very enjoyable, because I hate adds, especially if they move, blink or spout sounds at me:

Nuke Anything Enhanced: Like most Add-Ons this is a JavaScript based thingy, basically simple and small, with just one feature that may save most of the internet-as-is for you: It can remove - make invisible - almost anything on any site in two clicks. Delicious!!

Having mentioned Javascript, which I will say more about in the next section, you should realize that most websites, and nearly all commercial ones, now run some sort of script, that enable the makers  of these pages to do all manner of things, such as recording your presence, popping up ads, redirecting your browser, and God knows what else.

NoScript is an Add-On, in JavaScript, that allows you to block scripts comprehensively or selectively. This is also worth adding to Firefox, if only to protect yourself and have an idea of what may be going on, and as with AdBlock Plus the (probably unavoidable) setback is that to use it sensibly you need to have some basic knowledge about html and indeed JavaScript.

Possibly this last is not necessary, but it surely helps, and in any case it makes a lot of sense not to allow script at all, except on sites you trust - and please note that JavaScript is powerful and normally it is hard to tell what a script does, and indeed often not easy to get to see all the code it runs (since the code tend to reside elsewhere somewhere, and not in the page they operate on).

3. JavaScript:

I'd seen JavaScript before and indeed programmed a little in it and found it a sensible and clear language, which is something I cannot say about Java, that I don't like.

As it happens, JavaScript has actually nothing to do with Java itself, and is only called "JavaScript" because in the days it started, some 10 years ago, this seemed a convenient advertisement-policy, and Sun insisted on it.

As it is at present, it is a powerful language, with many facilities, which nevertheless is a scripting language because it is meant to add things and capacities to webbrowsers, and especially to Firefox. (There is a Lite version that works also on MS Explorer, but MS doesn't like open source and therefore developed its own JScript, about which I only know that it crashed my computer in a horrible and major way once, making me loose over 50 MB of data on my harddisk.)

If you want to, you can find out a lot about JavaScript by reading Mozilla's documentation (Mozilla is the organization behind Firefox and related products); google the net (or whatever search-engine you prefer) for "JavaScript Tutorials"; and help yourself a lot, if you are interested at all, by once more consulting the Add-Ons for Firefox.

Here there are quite a few utilities, editors and so on to see and write JavaScript code in. Here are four that I have used some and found something to like in, from fairly simple to quite sophisticated:

Phoenix: This is a basic editor that seems to work well and is helpful and useful.

JavaScript Debugger: This a quite a bit fancier than Phoenix, and it seems decent, except that the guys who made it - it seems to go by the name Venkman, internally - included a script that connects you to their site when you start it.

This is a reason I have used it very little, for I am against this on principle, just as I am against the recently evolved fashion to force you to go on line to see helpfiles for programs, probably because this adds ticks to those sites, which in turn adds financial gravy to the ads that usually are on it.

Apart from that, it appears like a decent programming environment.

Firebug: This is fancier and more recent than JavaScript Debugger and it is smart and useful - except that you do need to know some JavaScript.

Sofar, this seems to give the best and the most information about JavaScript Code, and the one niggle I have is that here too to see the Helpfile you have to visit the site of the makers.

There are at least two reasons why JavaScript is interesting for me, namely (1) that it is an object-oriented language, and I have spend rather a lot of time on that in the context of learning Smalltalk, that originated much of OOP (Object Oriented Programming) and (2) that it does give a part of what Smalltalk gives - that has the serious setback that it depends on its own environment, and runs on your computer more or less as its own OS (Operating System) - and not in a Smalltalk environment but in the environment of a browser (Firefox, specifically).

This gives it - in principle at least - a lot of power, and also in code that only requires a browser, and not a whole Smalltalk-environment. This has lots of advantages, and one of these is that there is a lot of code for JavaScript on the internet, and some of it is nicely bundled:

Mochikit: This is one of such bundles of code designed to make JavaScript easy and useful. As the makers say: "MochiKit - makes JavaScript suck a bit less", and indeed it does. You can download it for free from the last link, and it installs perfectly easy and well, and I found it very helpful. Also, it comes with excellent documentation.

4. Smalltalk news:

I am still following the Smalltalk world - see the index of BAP for more - basically because (1) it does work, once you've mastered Smalltalk and the environment in which it works (that varies some, or a lot, with the implementation you use); (2) I like the language; and (3) most of the people who program in it are helpful and intelligent, if for my tastes perhaps too much wedded to one language and one model for computing.

So here are some updates on some implementations of Smalltalk and some news related to Smalltalk:

Squeak is still struggling on, and the latest innovation is that there now is a trunk-image, which is related to a coding repository for developers.

The up side is that since this happened, around a month ago or so, lots of code has been added, and the down side is that I doubt it was a good idea - while it just doesn't work for me.

It does not seem to me to be a good idea because it was imposed by Andreas Raab with hardly any discussion on the Squeak developers list, quite probably because he was justifiedly upset about the slow rate of development of Squeak (and he knows a vast amount about Squeak, indeed wrote a lot of it, and has a firm that sells products based on it), but this seems to me to be just too centralized and undiscussed a move.

Now this is as may be - and something had to be done to pull Squeak out of the mire, and maybe I am a bit too pessimistic - but the real flipside for me is that I have now downloaded 4 trunk-images thus produced, and they don't even start. (Instead, they tell me I ought to drop the Squeak-image on the Squeak-executable, which I have been doing since July 2001, and also here: No juice at all.)

For me, this is an effective showstopper - but I admit there is a smallish chance it is my computer, and not the trunk-image, although the rest of the Smalltalks on my harddisk, including earlier Squeaks, all do as they are supposed to.

So let's consider the brotherly competition to Squeak:

Pharo is a fork of Squeak, because a number of developers (especially French or Swiss), some of them quite important also, were fed up with the slowness of Squeak's development.

It is mostly Squeak 3.9 (I think), except that some of the stuff that is in Squeak has been ripped from Pharo (that aims at a commercially useful Smalltalk, among other things); a goodly amount of the code seems to have been cleaned up some; the Pharo-interface is a lot nicer than that of Squeak; and otherwise it just works as I am used to in Smalltalk and Squeak.

The only problem I had with it is that it does have a working update system (as Squeak used to have), that I tried to use to update the distribution-image of Seaside, of which more below, that is mostly by people also doing stuff for Pharo, but that turned out to take ages and to be in eternal loops. (I may be mistaken, but sofar as I can see, after some 40 minutes of downloading and installing, at most 10% of what was to be installed was installed, and I saw the same thing occur again and again in terms of downloading and installing so I stopped.)

Seaside is another fork of Squeak and is meant for web-development. I don't know much about it, but I downloaded it with the result in the previous paragraph.

What is good about it, though, and why I downloaded it to take a closer look at it (which makes sense only if you know already how to program in Smalltalk, I think), is that there has been a recent free e-book written for it, that seems to be fairly well done, and is indeed what nearly all Smalltalk-environments need: Good, clear documentation.

I don't know whether Seaside is interesting for me, for I tend to believe that JavaScript will be more useful for my purposes as they relate to webdevelopment, among other things because unlike all Smalltalks it is not effectively enclosed in its own OS, that does not have all the power of Windows API, though it does have nice things plus the sourcecode of everything.

But it is a Smalltalk and it looked decent and comes with what appears reasonable and extensive documentation, which is why I mention it here.

5. Open source

I am much in favour of open source - and indeed all I reviewed in this text is open source, but I have lately been fairly pessimistic about its chances to oust or tame Microsoft, Apple, Sun, Google and other hidden source commercial players (though these differ also, and I am only for shortness's sake bunching them here together).

In any case: Firefox and JavaScript are good examples of what good open source may be and can do, for it is all open source and it is considerably better than what the commercial competition has to offer, viz. especially MS IE and JScript, these being Microsoft's browser adapted and commercialized from Netscape, and Microsoft's adaptation of JavaScript for its own purposes, while mangling it to fit in its vision of what computing should be (commercial, profitable, hidden source).

So that's fairly good news: It is possible to do better than Microsoft, also with regards to its flagship product (apart from Windows), and it has resulted in a better browser with fine Add-Ons.

Indeed, what pleased me most in all of the above - because I just hate ads: to me it is literally breaking into my own privacy, intentionally so, which I find despicable no matter what lousy justification smoothies may find for it - is the brief, short, powerful and effective Nuke Anything Enhanced.

For this makes browsing the internet far less irritating than it is when one can't block ads, that to me are spam to the power of spam. And this leads me to my last topic of this discursive BitsAndPieces:

6. A new internet is needed + 3 special reasons why

It seems to me highly desirable to split up the internet as is in a commercial net and a public net, with the commercial players, the advertisers, the commercial-Flash-on-you-pukers, and indeed also all that is commercial and more or less decent and justified, from banking to plumbing, on one net, and non-commercial content, varying from private persons websites, to schools and universities, the Wikipedia and much other educational matter (Stanford Encyclopedia, Victorian Net, Perseus, you name it...), on another net.

In fact, the split between commercial and non-commercial can be drawn quite easily, while allowing for overlaps: It is mostly that between open en hidden source and apart from that depends on whether makers of web-content want to cash in on advertisments and/or have web-pages in order to support a commercial product.

Next, the technological change should be easy and small, in principle, for it is a matter of server and storage capacity and processor speeds: All that seems really required is, next to a WWW a WCW, say: A Worldwide Commercial Web, all run on the same known principles as the WWW, with the same protocols a.s.o., except that on that WCW the user knows that the content is commercial and meant to be commercial, and indeed may very well be hidden source.

Since the server and storage capacities these days are great, as are the processor speeds, it seems this is technically quite feasible - and indeed in the interest of virtually all private persons, while also giving in principle new possibilities, on that WCW, to use all manner of hidden source pimped software to make it better looking and produce even livelier ads for the masses.

Besides, such a WCW serves a legitimate purpose: It is like a computerized set of Yellow Pages in the phonebooks of yore (just as the ordinary phonebook lists mostly private persons, and without ads).

And here are three final arguments:

First - it is and ought to be public and open: the internet has grown out of the efforts of private persons and people working at universities, and indeed out of what was from the beginning open source - Microsoft jumped in only for the money, and since then attempted to redesign it for its own needs.

Second - it makes solid sense:, the split I advocate - into a WWW and a WCW, to reduce it to acronyms - corresponds quite closely to a manifold split of the following kind

open source   - hidden source
personal        - commercial     
educational    - profit-oriented
individual       - organizational

Third - it may save the public media: Seeing what is happening to the classical papers (getting extinct fast, to be replaced by Foxnews and the like, at least for the masses) and to the media (mostly moronified to please the masses of pinheads and the smart-alecks who live off these pinheads), this could be an excellent means to regain something like the classic media, but on the net:

Make the media - in part, for there is also place for commercial media, but I am talking in fact of the major public benefits of real quality papers, produced by highly competent and trained journalists, preferably with a university education - into educational institutions, somewhat like universities are (in Europe), that is, mostly funded by public money and on a BBC-model.

Note that here too I am talking about known models, while I am not proposing anything requiring much innovation or investigation:

The BBC works and produces the best journalism to be had, without ads, and without state censorhip, so this ought to be possible as well for quality papers, on the same principles, largely for the same reasons also, except that computing, the commercialization of the internet, and the imminent collapse of quality journalism, have made this both more urgent and more easily possible.

Indeed, one may well combine such new quality media on the net with universities or polytechnics, whether special or part of the old universities, and make these into professional Media-Labs, as it were, working for the public, funded by public money, part of the educational institutions, and with their own (almost) ready-made schools for internet journalism there.

This seems to me the best way to go, to preserve the quality media, albeit in a new form; to give the internet back to the public, who created it in the first place; and to clearly separate what is commercial from what is not, and should not have to compete with, crowded out, or suffocated by, commercial players with private ends and megabucks of money.

And I know the above is a brief argument, that may be butressed and countered in quite a few ways, but it seems to me the only way I know of to keep the internet public, educational, open source and for and by private individuals, and also to retain quality journalism and quality media in the interest of all

(*) They're not alone in this, by the way: Part of their "service" is a licensed McAfee, which I have come to deeply hate, and strongly disrecommend: Slow, stupid, and not working in many respects - I can't even download an English version, and have tried 10 times now (I must be living in Holland, and hence must be a moron who can't read English, it seems is the reasoning behind it); I can't let it defragment my disk without locking up the computer; the moronified interface is slow as treacle and seems to hide rather than give information; useful utilities there are none and so on, and so on: No way this works, as most things in the McAfee suite only seem "to work" and the whole suit seems to be designed for the lowly educated of great faith and small brain.

Maarten Maartensz


        home - index - top -