Fire Destroys Southampton Fibre-Optics Center

Sam Haine '95 writes "BBC News reports that a fire has burnt down a CS facility at the University of Southampton. It's notable because the facility was one of the best in the world." From the article: "Some of the most advanced research work in the country, and indeed the world was carried out in this facility ... We probably will have to start from scratch, and it will take a couple of years to rebuild the facility"

Comments

That is certainly unfortunate.

[Sheetrock]

The only two things that I can think of that might be of some consolation are that because this dealt with technology much of the research should be in electronic form and backed up and that many times you'll discover a more efficient way of doing things when you go back and design the same thing a second time (although one normally does not have the luxury/misfortune to do so).

Backups?

[(negative video)]

What about the building's fire sprinkler system? Why did it fail? Or why didn't it have one?

Re:Backups?

[NoGuffCheck]

No need to worry, luckly the sprinkler system was off site so it survived the fire unharmed... phew! y,know, its close scares like this that make you really appreciate your plumbing.

Who dun it?

[TheDracle]

First Google buys all unused fiber optics:
http://hardware.slashdot.org/article.pl?sid=05/08/ 28/2156233&tid=217&tid=230&tid=193

To corner the market.

And now mysterious fires ravage the competition.

Worst case scenario more like couple of decades

[MMaestro]

More like couple of decades worst case scenario. Even assuming they backed up all their data off-site, its gonna take years to clean up and build a new building (~1 year for cleaning, estimate about 2~5 years to rebuild and bring the whole thing back up to speed). On top of construction, theres the loss of the machines, facilities and temporary unability to do work elsewhere for sometime (can't replicate that kind of research in your average college computer lab.) Theres also the now temporarily work-less researchers, the obvious political fallout and having to figure out how the fire started in the first place. Theres also the matter of who's gonna foot the bill for cleaning, construction and replacing all the lost material. On top of ALL that, I doubt theres a magic 'put everything back the way it was before button' on the backup servers which can instantly bring all the data back for researchers quickly and easily.

Combined, you're looking at an easy 5 years lost research time best case scenario. Worst case scenario you're looking at anywhere between 10~30 years lost time since some scientists may not want to wait for the facilities to be rebuilt and just take their expertise elsewhere and their not the sort you can replace easily. Theres always the distant (but unlikely) possibility, that they might not even rebuild the facilties and simply shelf or sell off the data to others.

And of course, this doesn't even touch the financial costs, the damage to the school's prestige and damage to the school's pride.

Firewall

[Kyeetza]

A massive fire has destroyed a leading computer science research facility.... They should've invested in a better firewall...

Sprinklers wouldn't have

[jd]

But halon gas fire suppression systems should have - at least, to the degree of it not getting totally out of control. Without oxygen, fires don't generally do a whole lot - halon largely works by displacing all of the oxygen, leaving the fire nothing to work with.

There's also the question of cannisters exploding... Cannisters generally don't do this - they tend to be rather boring, not even speaking much, unless there's something already happening. Cannisters will react to heat - but, like I said, a halon system should have dealt with heat sources long before they became a threat. Cannisters with explosive gasses CAN explode if the valve is leaky and there is a static discharge. But anyone leaving highly explosive substances around massive sources of static, or indeed, in containers that are faulty - well, they should expect something like this. You should generally store cannisters and gas cylinders in well-ventillated but secure locations containing no combustible materials or materials likely to pick up a static charge.

In practice, you can't go around stowing every single piece of equiptment in absolutely ideal conditions. In consequence, accidents like this are going to happen. Because they are going to happen, the important thing is to keep the impact to a minimum. A lot of effort over the years has gone, not only in building fire suppressing systems, but also in figuring out how to build structures that will contain a fire. The slower a fire can spread, the more likely it is to exhaust fuel and/or oxygen before it can find more.

Now, explosions get more problematic. Once you get explosions, there's not a whole lot even the best design can do, because you have to assume that there will be a sizable area affected. Aside from minimizing risk (through correct handling and operating procedurea) and trapping precursors (such as nearby fires, static, etc), there's not much that can be done. If you want to have a building survive explosions, you've got to design it very differently - lots of honeycombed structures that can absorb the high energies involved, for example. On the whole, though, you wouldn't design a fibre optics centre that way. Fibre isn't known for exploding. Fireworks factories SHOULD be built that way, and a lot of people killed in such explosions might well be alive if such buildings WERE built correctly for the conditions, but that's a whole different ball-game.