I thought that was a tad unfair. 1.3 has 'served' us well, and I raise a glass to it.
After all, wasn't the name "Apache" from 'A patchy' web server?
C'm on, Gavin. Credit where credit's due.
An increasingly creaky version of the web's most popular web server has finally been retired after twelve years serving billions of pages. The Apache Software Foundation has released HTTP Sever 1.3.42, saying this is the final release of the 1.3 branch, and there will be no more updates except for critical security fixes. …
Apache 1.3 has had ***-all developer interest for years. But the users kept on wanting support and updates. Result: an effectively-moribund but still widely-used server.
Making it official should help end this situation. Anyone who still wants to use 1.3 now knows officially they should go elsewhere (and pay) if they want further support.
Would be interesting to see if netcraft(or someone) had the data as far as approx how many sites out there were still on 1.3.x.
Myself I migrated my main personal server to 2.2 a couple of years ago(and I thought I was taking my time), the upgrade was easier than I was fearing, was quite painless. Made going from cyrus IMAP 1.x to 2.x seem like a walk across the street to work.
The other personal server I upgraded a few months ago and it was even easier, less custom configurations in it.
The reason I suffered a few day's pain upgrading from Apache 1.3 to 2.2. last summer was because 1.3 was no longer security supported on the newer Lenny Debian stable and I had to upgrade to Lenny from Etch. My Apache 1.3 configuration was so convoluted after many years of accretions that it took me a few day's work to unpick it, and I had to put various of the applications and site areas served into separate virtual domains with URL redirects. Things are much more modular and maintainable now that I've done all that. Moved on at the right time for me and the sites I support, but 1.3 served very well for many years with few if any problems.
Goodbye old friend youve "served" us well and saved us from IIS and the oracle web server (am I the only one who remembers that and its amazing perl cartridge stability (sarcasm...).)
Ill raise a glass of beer for you tonight.
I can hear the sound of some embedded os people spluttering in the distance mind :)
This is a song about a couple of adult people who have spent, oh, quite a long time
together, till one day one of 'em gets tired of patching and decides to leave.
Whether it's the patcher or the patchee who left is unimportant.
It's a breakup.
The lovely marriage of code and support is not over though. It just has a different colour coat.
While the code running an old application may not change, its environment does. New types of security threats emerge from sources and vectors unknown at the time the code was created, for example.
Your organic metaphor is quite apt, in fact. The ecosystem of the Internet is of greater complexity than any one person or group can predict; that is precisely why the Internet is such a fascinating thing to study... and so is biology, and the Universe itself.
When systems reach a certain level of complexity, subtle interactions between members of the set begin to make significant effects. For example, in a vat of chemicals with an energy source, a small proportion of species will interact in a catalytic fashion, creating new species. Some of these will react in a new way with other species; and in this way complexity grows exponentially. Ultimately you have seemingly transcendental phenomena like life appearing; but it's not magic, it's chaos with power input.
Software systems behave analogously to chemicals in the soup. The code is genetic material; the power input is our intellectual work in crafting software and systems; mutations are code bugs; and catalytic reactions are un-designed behaviour in software resulting from unforeseen conditions and a changing environment and infrastructure.
Software that is used in a changing environment without developer input is moribund. This is analogous to a species in an evolutionary dead-end. A changing environment will increase predation, while it is unable to change to compete. Decline and extinction, or, as you put it, "rot", are inevitable.
Biting the hand that feeds IT © 1998–2019