[WEB4LIB] Moldy old servers

[Brian Kelly]

Hi Thomas
      These figures are very interesting.  Have you thought of
documenting them to preserve a snapshot, to help spot any trends?
      We have been involved in such work at UKOLN, monitoring trends
across UK University (and more recently public library) communities.
      For example we've looked at trends in the use of search engines -
see latest figures at
http://www.ukoln.ac.uk/web-focus/surveys/uk-he-search-engines/survey-jan
2000.html
    We've noticed similar areas of concern e.g. use of very dated
software, software which is no longer available (migration problems if
it stops working), software with security concerns, etc.
     Note that in your list of problems with continued use of old
versions of server software you should add lack of HTTP/1.1 support and
poor support for caching - which adversely affects the network and
should be a concern for everyone.

Brian
--------------------------------------------------------------------
Brian Kelly, UK Web Focus
UKOLN, University of Bath, BATH, England, BA2 7AY
Email:  b.kelly at ukoln.ac.uk     URL:    http://www.ukoln.ac.uk/
Homepage: http://www.ukoln.ac.uk/ukoln/staff/b.kelly.html
Phone:  01225 323943            FAX:   01225 826838
----- Original Message -----
From: "Thomas Dowling" <tdowling at ohiolink.edu>
To: "Multiple recipients of list" <web4lib at webjunction.org>
Sent: Wednesday, January 26, 2000 10:49 PM
Subject: [WEB4LIB] Moldy old servers


> I've been doing some link checking, and you don't want to get me
started
> on the mind-numbingly stupid URL changes people foist on their users
> (hands up everyone who changes the circ desk's phone number and
> disconnects the old one everytime the library's place on the org chart
> changes.  Anyone?).
>
> In the course of going through HTTP headers for bad status messages, I
> gradually became aware of how many badly out of date servers I was
seeing.
> A little grep, a little sed, and here's the sorry breakdown:
>
>
> Out of 1903 servers (most of the North American library servers I know
> about):
>
> Apache, all versions: 696
> Apache, pre-1.3: 193 (28% of Apache installations)
>
> CERN httpd (development ceased, Summer 1996): 30
>
> IIS: 607
> IIS, pre-4.0: 82 (14% of IIS installations)
>
> NCSA (development ceased, Fall 1996): 36
>
> Netscape Enterprise: 276
> Netscape Enterprise, pre-3.5: 68 (25%)
> Netscape FastTrack: 25
> Netscape FastTrack, pre 3.0: 10 (40%)
>
> WebSTAR: 38
> WebSTAR, pre-4.0: 19 (50%)
>
> WebSite/WebSite Pro: 47
> WebSite, pre-2.0: 18 (38%)
>
> Dishonorable mention to the site running Apache 0.8.8 (that's a
college's
> main web server, folks), the 10 sites running beta versions of Apache
1.2,
> and the two sites running IIS 1.0.  Kudos to the 148 Apache sites that
are
> on 1.3.9 or 1.3.11.
>
> Obviously, there's no requirement to have an absolutely current
server,
> and I suppose these sites would upgrade if something were out-and-out
> broken.  But it bears pointing out that software developers don't just
> upgrade their products for the fun of it.  They fix bugs, they improve
> performance and stability, they add helpful new features, they plug
> security holes--for the most part, new versions are better than old
ones.
> Also for the most part, new versions fit even a chintzy library's web
> management budget--usually by virtue of being free--and it usually
isn't
> that big a job to install the upgrade.  We have some truly whacky
Apache
> customizations, and it still doesn't take me more than an hour to
upgrade,
> including downloading, compiling, configuring,and installing it.
>
> So what's the excuse for being *so* far behind?
>
> Thomas Dowling
> OhioLINK - Ohio Library and Information Network
> tdowling at ohiolink.edu
>
>
>