Slashdot Mirror
Maddog on "The Economics of Linux"
MojoT sent us a
link to Jon "Maddog" Hall's bit The Economics of Linux
that appears in Performance Computing. Talks about why
Linux is taking off and much more.
← Back to Stories (view on slashdot.org)
← Back to Stories (view on slashdot.org)
Right after this article came out in dead trees format I sent Jon a note (it still seems weird that he's a Compaq employee but he still has the dec address :)
In brief, it seems surprising that nobody in the trade press noticed until six months ago that Linux has more users now than Windows did in 1991 (yet Windows dominated the news pages of PC Week, InfoWorld and so on from 1990 onwards).
Two factors are of interest here. One is that Linux is not really a corporate desktop thing (yet), as Windows was at first (mostly because most home users couldn't afford the hardware in 1990-91; later with the fall in memory, disk and CPU prices, the release of W95 and the takeoff of the Web it zoomed madly in the home market). Linux has been quietly off in departmental corners serving print jobs and Web pages and email.
The second factor is that Linux has been thoroughly international from the get go. Thus its 10 or 20 or more million users have been spread far more widely than the concentrated corporate office visibility of Windows in the early years. In addition the localization that is key to expansion in non-US markets is very advanced under Linux. The market can grow UP, not OUT, from its current status. That is a crucial consideration for the future development trajectory of Linux.
My conclusion is that Linux is in a *stronger* position than many people realize to make further gains in departmental computing. There will be an inevitable backlash against Linux as potential users start to wrestle with its unfamiliar approach. But, mark my words, they will be assimilated. Why? Performance, price, features, reliability. Nuff said.
------
Bill Gates Is My Evil Twin.
My spin on it is this: If the linux install base is equal to the commercial unix install base then the independent software vendors Have to port to linux. No question.
Now desktop applications are a different story. Where one big (solaris|AIX|SCO|HPUX) box exists, there are 50 pc's. Running microsoftware.
We straddle the two worlds: datacentre and desktop. I run a linux desktop and use a linux server, as well as SCO and Solaris servers. What needs to happen is
Fully installed OEM systems.
Full desktop documentation, on said OEM systems
Someone needs to write a scripted Auto tuning tool, that can fix up the open file limit, the ulimits and such for big servers (so that catastrophe at MindCraft won't happen again)
A full "Compatibility Suite" of office productivity software. So that MS Office addicts can double click on a word document in their email and have (wordperfect|applix|staroffice) open it and convert it properly. And double click on a url and have (netscape|opera|lynx|whatever) open it properly. This ALMOST there people.
Its so close I can taste it. Gnome and KDE are very close. I use both. The productivity apps and the drag/drop interactivity is getting close to perfect.
we need someone to put together this type of office system, get it fully installed on an OEM box and ship it to Jesse Berst, the guy at Salon, and any other so-called technology writer.
-- Perl Hack, Web Hack, SQL Hack, Guitar Hack
Not to get down on maddog or anything, because this is really a pretty good expression of some concerns about Linux adoption -- but haven't we seen an awful lot of these recently?
It seems everyone is saying roughly the same thing:
1) The adoption of Linux-based OSes in a wider market is Good, provided it dilutes neither the freedom nor the hacker/geek appeal of the system.
2) Microsoft Office for Linux would cause wider Linux adoption, but this adoption would be too fast and reckless, and would damage both the freedom of the system (due to embrace-and-extend) and the cluefulness of the user base. A MS Linux distribution would be the same, but worse.
3) PHBs generally do not understand that the free-software model and the general attitude of openness and free exchange of information are what make Linux-based systems good, and if given the chance, may take steps that would damage this good.
To these commonly-observed points I would like to add a few, which I think have not been adequately addressed:
1) The Unintentional Hazard of Red Hat Linux
Red Hat may pose an unintentional hazard. While they clearly mean well, it is obvious by now that the mass media pay a great deal more attention to Red Hat than to other distributions, and that Red Hat, for tolerably obvious reasons, has no desire to change this fact. When they do notice other distros, the media tend to notice only other commercial distros, such as Caldera OpenLinux, and to ignore noncommercial ones such as Debian GNU/Linux and Slackware, even though Debian is presently the second-most-popular distribution, and has several arguable superiorities.
Red Hat may prove a hazard in that new programs, especially commercial applications, may be released in forms only suitable for use on Red Hat Linux. Besides being issued in RPM form -- which is not too much of a problem, but is an inconvenience, especially for new users -- they may be tailored to the set of libraries and other features present in the current Red Hat release, and possibly without testing for compatibility with other Linux-based operating systems.
In the case of free software releases this is not likely to pose a problem, because (for instance) the Debian maintainers could easily repackage the software in a Debian-compatible form. However, in the case of proprietary software, it is a serious problem, as proprietary-software houses are not as a rule interested in others repackaging their releases. While heavily political free-software advocates may not care whether proprietary software is available for their favorite noncommercial distro, such differences can clearly have an effect on the supportable user base of each these distributions.
2) Again, the Security Issue
As I have noted before, one issue critical to the good performance of Linux-based systems is security. As evidenced by the frequency with which newbie Linux users are rooted or otherwise cracked, security is being traded off for ease of use. This Must Stop.
We should expect that Linux-based systems will get easier to use, simply because so many people are now putting effort into that direction -- see the various documentation efforts, as well as the LUIGUI project. What we must be sure of, however, is that security is not sacrificed in this pursuit. Newbies and end-users need more help, not less, in securing their systems than do advanced users or sysadmins. Distros need to have default configurations that favor network security over network openness -- ftpd and the like being disabled or limited by default; ssh or SSLTelnet being preferred over Telnet; etc.
Another element of this point is that updating is one function which also needs to be made accessible to the end-user and the newbie. Keeping one's system up to date is an essential part of protecting it from attacks. One of our boasting-points is the fact that security patches for Linux systems are available quicker than patches for, say, NT -- but if these patches are not applied, the effort is wasted. In my mind, Debian has done the best job in making updating easy -- apt-get update ; apt-get upgrade is all the user need do, and even this could in theory be automated further. (One possibility would be to have an option which updates only packages whose installation requires no user interaction, and running this update every day in cron.daily.)
In any case, security must not be neglected in favor of wider adoption or ease of use, or the benefits of Linux-based systems will be severely hampered.
3) The Hacker/Geek Arrogance Factor.
Several people here have expressed views in common with Pablo Averbuj's Letter to Debian about Friendliness, whose "Executive Summary" read as follows:
1. Stupid Users are Bad.
2. Stupid Users are Bad for Debian.
therefore:
3. Stupid Users should be ignored.
While it's not fair at all to dismiss M. Averbuj's position as one of simple hacker/geek arrogance, there has been no end of arrogant hackers and geeks who have rallied behind a very similar position: namely than Linux is For Geeks Only, and that no effort should be made to popularize it.
The problem I have with this position is that it leads its followers to marginalize themselves, and I suspect that if followed long enough and in the face of mounting popularization, it will lead to the formation of a faction with a bunker mentality: self-identified "hardcore" Linux hackers who want nothing to do with anything like ease of use or popularization, and who decry all such efforts as misuse or misappropriation of the True Hackerly Linux.
Such an exodus, of course, would only reduce the cluefulness of any popularized Linux. It would also lead to more bogus political infighting of the kind we already see too much of. This would be Bad, not only because it would make us all look like flamers, but because it would reduce the amount of productive cooperation in the whole effort.