Then the conclusion is even easier :
"Redhat's (mostly business) users don't need or just don't care about disk compression so they don't ask for it"

Disk compression is a terrible band-aid for a cheap and easy-to-fix problem. Maybe it's time people rethink their storage strategies first...

Win2k was probably Microsoft's best piece of engineering... too bad it's almost impossible to run it on modern hardware.

As for Linux 3 click LDAP/AD join : All RedHat-deratives have had that for a long time if you were OK with plain LDAP. The system-config-autentication package sorted you out for LDAP and NIS. Both in the GUI or in a curses interface.

It SHOULD have been... unfortunately they don't care about the quality of an LTS release anymore than they do of a normal release. Another big reason for switching every machine I can over to a proper enterprise class distro, Scientific Linux in my case.

Ubuntu is NOT a proper server OS, no matter how many fanboys may like it... get a proper enterprise distro (RHEL or clone, or SLES) or get Debian to be sure you get timely fixes. Ubuntu is an obvious example of how NOT to treat 'enterprise' users; just look at the tons of bug reports still open for 10.04 (which IS the latest LTS release) ; many of the the reports contain trivial fixes but they hardly ever get pushed. It may be a nice desktop OS but IMHO it has no place on a server.

PostgreSQL might have been better engineered, but running, and maintaining, a MySQL installation is a lot easier; when working at an ISP I maintained dozens of both and always preferred MySQL. Restoring a backup? MySQL is a LOT easier and needs less work. Making a special backup? MySQL is a lot easier. Adding a user and granting rights? I prefer MySQL's way of doing. (I'm assuming all commandline here... maybe if you used a GUI things would work out differently but he's talking server and LAMP, which in my book means no GUI)

Only when you need the extra features of PostgreSQL you should consider it, but for a simple webbased forum package it's very much overkill and only gives more maintenance grief, specially as the original poster doesn't appear to have much experience with either system and the PostgreSQL learning curve is a los sharper.

Obviously you're not Dutch...

Every big city has between like 5 and over a 100 websites, of which almost 50% nowadays uses SSL (which by itself is a good thing!) Things like social housing, requesting a new passport/drivers license, every city has their own website(s) and almost all are secured by SSL as all those things involve personal data.

I used to work for a big hosting company who hosted stuff for many bigger cities. I remember Amsterdam having over 4 dozen websites just running at our company, linked to hundreds of URL's, most of them equipped with SSL.

Getting a new certificate meant going through the proper chain of authorities... most of the people involved were government employed and (thus) not always well motived and sometimes just clueless as to the urgency. I've seen things like this take weeks. It's not like they actually let the two technical people on both ends just do their jobs, that's WAY too simple. You'll encounter probably half a dozen of auditing/managing layers at least.

Some cities at least were smart enough to get wildcard certs... which make these things a lot easier.

Is it a waiste? Dunno, don't really care either. The more I learned about our government's IT, the less I trust it. So I do everything the old-fashioned way by paper and snail-mail.

You obviously have no clue of all the steps involved...
Most sites are hosted externally, usually with 2-3 parties involved per site. You need to go through all those hosters change / support systems, which might take hours but can also easily take days (if not weeks....) Add in that it's still holiday season, the fact that the severity of the incident means that many politicians and public servants will want to have their piece of the actions and you have a recipe for a longwinded mess.

With cert renewal you can plan for this and if you start on time nobody notices anything. This however is totally unplanned and I would be VERY surprised if they manage to fix all sites within a week.

That might be true in some shoddy companies, but more and more companies have discovered that experience can't really be bought, let alone outsourced. People with an MBA might be able to talk the talk, but can produce nothing but useless masses of paper. In my environment I've not heard of a single good engineer who got laid off because he was too old; on the other hand I did hear about good engineers brought back from retirement because nobody could really replace them.... I very much doubt many MBA's get that option.

I'd get a second major in either :
Electrical Engineering ; sounds like a good allround option since hardly anything these days is purely mechanical anymore. I studied Mechatronics and still daily see the benefit from having a solid knowledge base in both diciplines.
Chemistry ; might be a good major too, for more insight with regard to material science, lubrication and propulsion (fuel types / batteries)

Maybe this will FINALLY mean no more nspluginwrapper crap;
Every time I try to open more than one PDF (Ubuntu 8.04 LTS / 10.04 LTS) nspluginwrapper goes nuts, no more PDF rendering....

And can we have good old usenet back? :)

But seriously, I never 'got' Facebook and Google+ doesn't appear to be any more useful.

Indeed; even the most basic 5.6 security updated were delayed way too long; this to me is a much bigger issue than delaying the X.0 release. I don't mind waiting for a week after release by RedHat, but months is way too long! And the way the CentOS dev's behaved on the mailing list didn't boost confidence much either...
Any .0 release needs serious testing and usually a patch round before it's production suitable anyway (at least both 4.0 and 5.0 weren't good enough to use in a production environment...)

I've been deploying only SL machines lately (which suit me just fine; I hardly use anything but the supplied RPM's and a few home-baked so no compatibility issues), but I'm left with one CentOS5 server and am seriously considering a migration... but I'll wait with that depending on how the 5.7 release plays out given the work and risks.

Interesting... are you on 5.1 then? We're 'stuck' on 5.0 but with the latest release (at least no newer version available through yum) The export to another cluster sounds awesome! Just had another look, it's not there in our version.

As for integration with SGE : I can see the queues; I just can't modify most of the queue properties.

Bright Cluster Manager is quite nice, but still lacking loads of things. The shell is powerful but very cryptic, the graphical interface doesn't allow certain operations to be done on a range of selected nodes for instance...

Also the 'integration' with for instance Sun Grid Engine (which is supported) is not very thorough (specially with regard to setting up queues, something the Sun tools already suck at) I still need Bright Cluster Manager + the SGE tools + Sun ARCo to get almost everything done, and at times it feels like a lot of duplicated effort and some good old *NIX handywork is still required to really get things moving.

However the development team is very receptive to constructive feedback and much has changed over the past 6 months.

We're not talking about the average desktop, where apparently loads of bling is 'needed' these days. A cluster is usually for processing loads and loads of data.
Most in-house software usually compiles just fine on CentOS/RHEL/Suse , getting the compiler suites to work might even be easier than under Ubuntu (10.04 with Intel V11.1 libstdc++5 anybody?) , much professional software isn't even supported under Ubuntu (and specially with scientific software such things DO matter)

Trying to claim Ubuntu is more secure because it's more recent shows nothing but ignorance.

What people want in an HPC environment is that it processes data; month after month, always with 100% predictable result and without downtime. A proper enterprise distro gives you that. The 0.01% performance gain in a HPC environment of a newer kernel doesn't make sense compared to the pains of having to roll your own updates because the distro maker stopped the updates. Just look at Ubuntu 8.04 LTS right now to see what I mean, compared to RHEL5 (which is older, but sees much more maintenance) And remember; you don't 'just upgrade' a cluster's distro after a year or two. The OS that you install after it's rolled in usually lasts for the lifetime of the cluster, or has to last for at least 2-3 years.