Corporate taxes are exactly the same as raising income tax, except you are paying at the point of purchase rather then the point of earning.
So the fees that the large accounting firms like KPMG charge to productize "lowering your corporate taxes" and all the expense for buying things like a LILO/SILO product... those are all free? And all the people rich enough to have most of their money in a "company" so they don't pay as much tax... that's just good for everyone?
The logical implication is that corporate tax should be zero, and be made up by income tax on the middle/lower classes, because of course everything they buy will then come down in price... riiiight?
actually you're full of shit, because I've seen plenty of fired programmers
I'd mostly agree with what mkcmkc says, and was the first thing I thought when I saw this article... and what you said doesn't contradict it. In my 15+ years in the industry I know I've seen 1 (one) person fired for not actually being competent. I've seen a lot of people fired for HR or "process" type violations, or because someone "high enough" in the mangement chain didn't like them. But that's hardly the same thing.
One place I worked at had a semi-official policy of hiring people in at really low wages and only giving them raises if they were competent (with the assumption that they'd then quit). Some startups with a burn rate I've been at had a semi-official policy of "least competent, first to go in the next round of layoffs".
The main reason seems to be that it's hard to fire people, even ignoring the mental effort of having to tell someone they don't have a job... doing it "properly", so you don't get sued, is a significant amount of real effort for management (even true when everyone is under "at will" contracts). I think the person in management also has to be pretty good a their job to even understand there is a problem to begin with, esp. unlikely if management is "programmers who moved up".
In fact probably the most appealing thought of starting my own company is that I'd never have to work with any of those "unfireables" again. Of course, I'd probably hate my boss then:).
we know that most people (sadly) are using some version of IE currently; ergo, if they install IE8 and it makes itself the default, this is good for a variety of reasons
What are you smoking? If they upgrade from IE6 to IE8, and IE6 is the default... they don't need to change any settings!
The only time doing/not-doing this matters is when the user has specifically set firefox/whatever to be the default instead. Of these users there will be the users who "know what they are doing", and will probably have done the non-automated install, so will either not experience this or set it back immediately. And there will be the users who had someone fix their computer for them (likely the much bigger group), and they will almost certainly just get forced into using IE8 just knowing that "everything is different".
It seems reasonable to assume that this action was done purely to force the later set of people to switch and thus. make their "using IE" and "people who love IE8" stats. better. This would be annoying, but tollerable, from a small developer. From a convicted monopolist people should go to jail, not that I think they will get punished.
arstechnica 2008 year end report in ipv4 implies that 4% is actually a pretty big chunk... not that I think GM/MIT/etc. are stupid enough to just give the IPs back.
You're right, Sun had no idea their new license would be incompatible with Linux because they wanted to be compatible instead of doing the slimy thing and trying to make it be a selling point over using Linux, which everyone was doing. Alas. for them RMS and Linus travelled back in time and created/used the GPL just to thwart poor Sun.
"Right, so all those times Sun took BSD code and closed it screwing over the users..."
Screwing over which users, as compared to what (fictional) alternative scenario? A BSD license says anyone can use the code anywhere.
You are arguing in (self invalidating) circles. The obvious alternative scenario where Sun/etc. has to write 100% of the code instead of only 5-10% but locking all of it from the community. And yes, obviously, the BSD license says other people can harm you with your own code... that was my point.
All I'm saying, is that contrary to the typical slashdot image of the selfish proprietary code authors hoarding their IP, I and other closed-source authors happily cooperate on some BSD code.
Fine. I'll even agree, you are wonderful and happily co-operate. But you are not everyone and your initial argument was that BSD is "better" because you can cooperate with the community using it, but not using the GPL. My reply pointed out that there is a negative side to this as well as the positive one, and that is the reason for the (L)GPL.
As an analogy, BSD is like always cooperating in the prisoners dilemma while the GPL is more like playing tit-for-tat... in theory you can do much better with the former, but in the real world it often doesn't work out so well for you.
So while in theory GPL requires quid-pro-quo contributions, in practice BSD gets more. At least from me.
Right, so all those times Sun took BSD code and closed it screwing over the users, doesn't count? Or SGI? Or BSDI. Or MicroSoft. Or the people who desperately wanted to do it with GCC, or with the Linux kernel. Or with Samba. Or...
Saying you will only use BSD code, and thus. only contribute to it is fine. You are obviously free to do so, but pretending the rest of the world (or even close to half) works the same way is just insane ravings.
Most high-power Linux developers have tended to put their money where their mouth is in previous similar situations. Despite the environment, I doubt they'd have trouble finding new jobs.
Which situations? Ones where the companies basically went out of business and they trickled to RH/SuSE/Sun/etc., and maybe a few to Intel or IBM. But let's pretend they'll all walk out if Oracle buys them... where do they go work now? SuSE just layed a bunch of people off (some of which are now working at RH), Canonical don't employ developers, Sun hasn't been top of the list for a while but the last month was the final nail in that coffin. Do they all go work at Apple on darwin?... will Apple hire them?
This may be nitpicking.
As great as linux is, its dependability is weakened by the hardware its running on...
Solaris gets its nines (seriously ridiculous uptimes) by the software/hardware combo: Solaris running on Solaris hardware.
Linux runs on bigger HW than Solaris does, including IBM mainframes and 1024+ proc. "boxes" from SGI. Sure, that doesn't mean the cheap Dell boxes are better than the expensive Sun ones... and maybe some people still have an impression of Sun providing the high end, but it's not been true for a long time now.
On a slightly more serious note, CentOS 5.2 users that are upgrading to 5.3 should do "yum update glibc" first before a "yum update" because of a yum locking issue.
If it's the issue I think you are talking about, it's an rpm/glibc locking issue... and only happens when you run rpm inside the scriptlets of packages (basically glibc upgraded the pthread locking, so you can't have old and new users at the same time). You're not wrong in the solution, but there's nothing yum could have done differently.
For us at work, CentOS 5.2 broke nss_ldap (and it's still broken with 5.3)
If it's been broken, for you, for that long. Then you really need to either fix it yourself, or pay RH for support and open a ticket with them.
I admit that I didn't get fallout 3 because I knew they were holding the DLC back. Then again, I'd happily recommend Bioshock and the ps3 exclusive DLC. In many ways the X-box vs. ps war is only screwing the gamers, on the other hand who wants there to be just one player (unless it's Nintendo;).
Retirement savings accounts (IRAs): 1) IRAs are given a lower interest rate. 2) When IRA CDs are renewed automatically, it is always at a lower interest rate, taking advantage of people who are extremely busy before the renewal date, and don't have time to research a better rate at a different bank.
An IRA is a type of account, mainly used for retirment savings due to how tax is done on it, you can have basically anything in it. A CD is a loan to a bank, much like a savings account but with a higher rate of interest due to limits on how you can move the money about.
You are either intentionally lying, or more likely are just repeating words you do not understand.
Savings accounts: 1) Banks advertise a high rate, then lower the rate later, with very little or no notice. 2) Accounts don't show the interest rate. 3) Banks advertise phrases like "Great Rate" and then give an especially poor rate
All normal savings accounts I've seen have a fixed rate, in the 0-1% range for "low" deposit amounts (this is not good for a long term investment). But you can also withdraw money from an ATM directly from the "savings" account (this is not possible with any long term. investment I know of).
I've never seen a bank that didn't show the interest rate prominently. The rate is often "great" when compared to logically similar offers. If I can buy a brand new Nissan for $1,000 or a Ferrari for $10,000, the later might be a much better deal... but that doesn't help if I only have $5,000.
Last I checked dpkg didn't sign packages (like rpm), and thus. apt didn't check those signatures. The repodata is signed though, which should still be secure using apt-p2p... AIUI.
Stop right there. Sun is one of the biggest corporate contributors to open source. Go ahead, count lines of code. I'm betting Sun will be in the top two if not #1.
Here's a brief list of things Sun has open sourced:
Solaris - Their entire OS, including ZFS and Dtrace
SPARC - Their CPU line
Java - Maybe you've heard of it.
OpenOffice - The office suite that ships with every desktop Linux distribution.
VirtualBox - A GPL desktop virtual machine.
NetBeans IDE - A multi-platform IDE.
OpenDS - LDAP Directory Server
High Availability Cluster
NFS
Solaris was only "released" after Linux had repeatedly shown Sun that being proprietary was a bad idea. And even now OpenSolaris != Solaris, plus you won't need both hands to count the number of external contributors.
SPARC, that's nice... you know many OpenSource developers who have chip fabrication plants?
Java... yeh, after IBM, Red Hat, SuSE, GNU, etc. spent 10 years reimplementing most of it. And Mono had mostly implemented C# 1.0. So they were basically forced to open it, or become as irrelivant as they are in the OS world.
OpenOffice... yeh, well done. They had a flash of insight (for once) that they couldn't possibly compete with it without open sourcing it.
Netbeans... again was proprietary, got shot in the face by eclipse, was reluctantly released (hmm, sounds familiar).
OpenDS - yeh, I know a lot of people using this instead of OpenLDAP or Fedora/RH directory server.
NFS - They made a single code drop at the begining, to make it a std., so they could sell "the good implementation" which was only in Solaris (and never released until decades later).
ZFS - Same deal, they've released it as narrowly as possible to try and make it a std.... and it looks like BTRFS is going to shoot them in the face, just like something did every other time.
So, sure, by lines of code dumped onto the 'net Sun are an open source company. In pretty much any other way of evaluating it though, they suck.
No, when all big newspapers are out of business people will have news from other sources. Perhaps a less concentrated, less politically biased model of news distribution and generation.
This is just wishful thinking to assume that random people will be less biased, although you might be able to argue that they will be biased in different ways... but that's not the same thing, and it's not necessarily better.
The concept that a media behemoth is needed to generate content is outdated, perhaps excepting movies, since you need big budgets for productions (but not for distribution anymore).
I think this is missing some info. too, both newspapers and movies have had monopolies on production and distribution. And as long as you aren't talking about digital distribution, both still do. The problem is that newspapers have crossed the point where digital distribution is often better and physical distribution, so all they have is the production monopoly... and they've been canibalizing their production budgets to keep their physical distribution monopolies going.
With TiVO, netflix, PSN and YouTube all trying to push digital distribution of movies/TV we are going to get to the crossing point soon (but, not there yet, IMO) where buying/renting DVDs isn't the "best" distribution anymore. I'd hope/assume the movie studios won't react by firing 90% of their actors/directors/etc.... but you never know.
It's good to see you didn't read the comments, where the Twitter admins/developers explain why the arguments in the blog aren't true. Meh, personally I hate Java and Ruby in equal amounts... but just saying "Ruby can scale" over and over, isn't a winning argument.
What about closing sockets that are only referenced in the poll() call of another thread?
Uhm the same thing that happens in Linux? poll() isn't unique to Linux, other OSes have it too.
Maybe this was too vague. If you close a socket that is only referenced in poll, then on Linux it behaves as if you closed the socket just after poll() returned (ie. nothing happens) and on FreeBSD IIRC it behaves as if you closed the socket just before you called poll (poll() call returns immediately with EINVAL for that fd).
This kind of thing would be almost impossible to make behave identically, under an emulated syscall. Nevermind, the comment by Matt Burke answered my question.
Several more correct options are: just produce decade-by-decade visualizations, or else produce year-by-year visualizations, but assign a "1960s" datapoint as a 1/10-weight datapoint in each of 1960 through 1969.
Maybe more correct. Averaging sucks x3, what about if 1966 had 2x as many events as the other years... what about if it only had those extra events for a reason (and so all events of that type were recorded with the year and not just the decade). I think really you have to either merge all the data to the same level (only give decade numbers), or distinguish the data in some way (so 11 data points when displaying data from 1960-1969 and "1960s").
FreeBSD has no problem running Linux binaries, linux binary compatibility has been there for years, I used it to run linux binaries that hadn't yet been ported to FBSD yet in 97, I still run several Linux binaries on my FBSD servers.
Does this actually work 100%? How?
And, yes, I understand you can do syscall emulation. But what about what happens behind the interfaces. For instance I find it hard to believe that TCP_CORK/mremap/epoll/etc. "works" when FreeBSD has refused (decided not to, whatever) to natively support it for years now. AFAIK FBSD doesn't have splice()/tee()/etc. either... do they hack some of this in userspace?
But even that seems like the easy stuff, what does FBSD do when I open("/proc/*") and start parsing stuff? What about closing sockets that are only referenced in the poll() call of another thread? Anything that hits the drivers "deeply" like X, pulseaudio, etc. seems like it'd be impossible to support. SELinux is just not going to work, probably dito. somewhat releated stuff like the audit interface / netlink (maybe that classifies as "deep" driver knowledge though).
Then there's the really crazy stuff where you have the same interfaces natively but they operate subtley differently in weird corner cases, DJB has a page on some stuff, but I doubt anyone sane enough to have commit privs. knows all of these (if anyone at all does).
This is highly misleading way of interpreting that data. Esp. given the caveats that lwn.net has to do just tracking the companies for linux kernel development. From what I can tell they unpacked all packages in Debian, and then assigned the "corporate contributor"... so all of OpenOffice and user space NFS goes to Sun, even though they did little more than a code dump, 10 years ago, in the later case.
In general I'd hope that "the community" is not using Solaris now because they see through BS stats. like the above, and know that Sun are much less of a participant in the community than Red Hat. But then, given the free ride Canonical gets sometimes, I do worry that it is just a case of network effects working against Sun.
Or not. The quote says "we are creeping towards $30 million" and that this would make Canonical "self-sustaining", for regular updates, which is their current model of very little development and a lot of marketing.
So, yeh, it's possible that "creeping towards" does actually mean "really close to, and getting closer every day" and not just "going from 1 to 2 million this year". But I bet against that, personally.
And who is paying that $30 mil, is it desktop people... or self supporting server users like the French govt. / Wikipedia. My guess is that any money is coming from the later.
So at the end of the creep towards $30 mil, what do you get? Still not enough money to fund actual development (feel free to log bugs and RFEs at bugzilla.redhat.com), and a focus on servers instead of the desktop for what little work they do (because that's who is paying).
Slashdot Mirror
So the fees that the large accounting firms like KPMG charge to productize "lowering your corporate taxes" and all the expense for buying things like a LILO/SILO product ... those are all free? And all the people rich enough to have most of their money in a "company" so they don't pay as much tax ... that's just good for everyone?
The logical implication is that corporate tax should be zero, and be made up by income tax on the middle/lower classes, because of course everything they buy will then come down in price ... riiiight?
I'd mostly agree with what mkcmkc says, and was the first thing I thought when I saw this article ... and what you said doesn't contradict it. In my 15+ years in the industry I know I've seen 1 (one) person fired for not actually being competent. I've seen a lot of people fired for HR or "process" type violations, or because someone "high enough" in the mangement chain didn't like them. But that's hardly the same thing.
One place I worked at had a semi-official policy of hiring people in at really low wages and only giving them raises if they were competent (with the assumption that they'd then quit). Some startups with a burn rate I've been at had a semi-official policy of "least competent, first to go in the next round of layoffs".
The main reason seems to be that it's hard to fire people, even ignoring the mental effort of having to tell someone they don't have a job ... doing it "properly", so you don't get sued, is a significant amount of real effort for management (even true when everyone is under "at will" contracts). I think the person in management also has to be pretty good a their job to even understand there is a problem to begin with, esp. unlikely if management is "programmers who moved up".
In fact probably the most appealing thought of starting my own company is that I'd never have to work with any of those "unfireables" again. Of course, I'd probably hate my boss then :).
What are you smoking? If they upgrade from IE6 to IE8, and IE6 is the default ... they don't need to change any settings!
The only time doing/not-doing this matters is when the user has specifically set firefox/whatever to be the default instead. Of these users there will be the users who "know what they are doing", and will probably have done the non-automated install, so will either not experience this or set it back immediately. And there will be the users who had someone fix their computer for them (likely the much bigger group), and they will almost certainly just get forced into using IE8 just knowing that "everything is different".
It seems reasonable to assume that this action was done purely to force the later set of people to switch and thus. make their "using IE" and "people who love IE8" stats. better. This would be annoying, but tollerable, from a small developer. From a convicted monopolist people should go to jail, not that I think they will get punished.
arstechnica 2008 year end report in ipv4 implies that 4% is actually a pretty big chunk ... not that I think GM/MIT/etc. are stupid enough to just give the IPs back.
You're right, Sun had no idea their new license would be incompatible with Linux because they wanted to be compatible instead of doing the slimy thing and trying to make it be a selling point over using Linux, which everyone was doing. Alas. for them RMS and Linus travelled back in time and created/used the GPL just to thwart poor Sun.
You are arguing in (self invalidating) circles. The obvious alternative scenario where Sun/etc. has to write 100% of the code instead of only 5-10% but locking all of it from the community. And yes, obviously, the BSD license says other people can harm you with your own code ... that was my point.
Fine. I'll even agree, you are wonderful and happily co-operate. But you are not everyone and your initial argument was that BSD is "better" because you can cooperate with the community using it, but not using the GPL. My reply pointed out that there is a negative side to this as well as the positive one, and that is the reason for the (L)GPL.
As an analogy, BSD is like always cooperating in the prisoners dilemma while the GPL is more like playing tit-for-tat ... in theory you can do much better with the former, but in the real world it often doesn't work out so well for you.
Right, so all those times Sun took BSD code and closed it screwing over the users, doesn't count? Or SGI? Or BSDI. Or MicroSoft. Or the people who desperately wanted to do it with GCC, or with the Linux kernel. Or with Samba. Or...
Saying you will only use BSD code, and thus. only contribute to it is fine. You are obviously free to do so, but pretending the rest of the world (or even close to half) works the same way is just insane ravings.
Which situations? Ones where the companies basically went out of business and they trickled to RH/SuSE/Sun/etc., and maybe a few to Intel or IBM. But let's pretend they'll all walk out if Oracle buys them ... where do they go work now? SuSE just layed a bunch of people off (some of which are now working at RH), Canonical don't employ developers, Sun hasn't been top of the list for a while but the last month was the final nail in that coffin. Do they all go work at Apple on darwin? ... will Apple hire them?
Linux runs on bigger HW than Solaris does, including IBM mainframes and 1024+ proc. "boxes" from SGI. Sure, that doesn't mean the cheap Dell boxes are better than the expensive Sun ones ... and maybe some people still have an impression of Sun providing the high end, but it's not been true for a long time now.
When Oracle customers started switching to Linux, DTrace didn't exist.
If it's the issue I think you are talking about, it's an rpm/glibc locking issue ... and only happens when you run rpm inside the scriptlets of packages (basically glibc upgraded the pthread locking, so you can't have old and new users at the same time). You're not wrong in the solution, but there's nothing yum could have done differently.
If it's been broken, for you, for that long. Then you really need to either fix it yourself, or pay RH for support and open a ticket with them.
I admit that I didn't get fallout 3 because I knew they were holding the DLC back. Then again, I'd happily recommend Bioshock and the ps3 exclusive DLC. In many ways the X-box vs. ps war is only screwing the gamers, on the other hand who wants there to be just one player (unless it's Nintendo ;).
An IRA is a type of account, mainly used for retirment savings due to how tax is done on it, you can have basically anything in it. A CD is a loan to a bank, much like a savings account but with a higher rate of interest due to limits on how you can move the money about.
You are either intentionally lying, or more likely are just repeating words you do not understand.
All normal savings accounts I've seen have a fixed rate, in the 0-1% range for "low" deposit amounts (this is not good for a long term investment). But you can also withdraw money from an ATM directly from the "savings" account (this is not possible with any long term. investment I know of).
I've never seen a bank that didn't show the interest rate prominently. The rate is often "great" when compared to logically similar offers. If I can buy a brand new Nissan for $1,000 or a Ferrari for $10,000, the later might be a much better deal ... but that doesn't help if I only have $5,000.
Last I checked dpkg didn't sign packages (like rpm), and thus. apt didn't check those signatures. The repodata is signed though, which should still be secure using apt-p2p ... AIUI.
Solaris was only "released" after Linux had repeatedly shown Sun that being proprietary was a bad idea. And even now OpenSolaris != Solaris, plus you won't need both hands to count the number of external contributors.
SPARC, that's nice ... you know many OpenSource developers who have chip fabrication plants?
Java ... yeh, after IBM, Red Hat, SuSE, GNU, etc. spent 10 years reimplementing most of it. And Mono had mostly implemented C# 1.0. So they were basically forced to open it, or become as irrelivant as they are in the OS world.
OpenOffice ... yeh, well done. They had a flash of insight (for once) that they couldn't possibly compete with it without open sourcing it.
Netbeans ... again was proprietary, got shot in the face by eclipse, was reluctantly released (hmm, sounds familiar).
OpenDS - yeh, I know a lot of people using this instead of OpenLDAP or Fedora/RH directory server.
NFS - They made a single code drop at the begining, to make it a std., so they could sell "the good implementation" which was only in Solaris (and never released until decades later).
ZFS - Same deal, they've released it as narrowly as possible to try and make it a std. ... and it looks like BTRFS is going to shoot them in the face, just like something did every other time.
So, sure, by lines of code dumped onto the 'net Sun are an open source company. In pretty much any other way of evaluating it though, they suck.
This is just wishful thinking to assume that random people will be less biased, although you might be able to argue that they will be biased in different ways ... but that's not the same thing, and it's not necessarily better.
I think this is missing some info. too, both newspapers and movies have had monopolies on production and distribution. And as long as you aren't talking about digital distribution, both still do. The problem is that newspapers have crossed the point where digital distribution is often better and physical distribution, so all they have is the production monopoly ... and they've been canibalizing their production budgets to keep their physical distribution monopolies going.
With TiVO, netflix, PSN and YouTube all trying to push digital distribution of movies/TV we are going to get to the crossing point soon (but, not there yet, IMO) where buying/renting DVDs isn't the "best" distribution anymore. I'd hope/assume the movie studios won't react by firing 90% of their actors/directors/etc. ... but you never know.
It's good to see you didn't read the comments, where the Twitter admins/developers explain why the arguments in the blog aren't true. Meh, personally I hate Java and Ruby in equal amounts ... but just saying "Ruby can scale" over and over, isn't a winning argument.
Fedora has come with working multilib for a long time now.
Maybe this was too vague. If you close a socket that is only referenced in poll, then on Linux it behaves as if you closed the socket just after poll() returned (ie. nothing happens) and on FreeBSD IIRC it behaves as if you closed the socket just before you called poll (poll() call returns immediately with EINVAL for that fd).
This kind of thing would be almost impossible to make behave identically, under an emulated syscall. Nevermind, the comment by Matt Burke answered my question.
Maybe more correct. Averaging sucks x3, what about if 1966 had 2x as many events as the other years ... what about if it only had those extra events for a reason (and so all events of that type were recorded with the year and not just the decade). I think really you have to either merge all the data to the same level (only give decade numbers), or distinguish the data in some way (so 11 data points when displaying data from 1960-1969 and "1960s").
Does this actually work 100%? How?
And, yes, I understand you can do syscall emulation. But what about what happens behind the interfaces. For instance I find it hard to believe that TCP_CORK/mremap/epoll/etc. "works" when FreeBSD has refused (decided not to, whatever) to natively support it for years now. AFAIK FBSD doesn't have splice()/tee()/etc. either ... do they hack some of this in userspace?
But even that seems like the easy stuff, what does FBSD do when I open("/proc/*") and start parsing stuff? What about closing sockets that are only referenced in the poll() call of another thread? Anything that hits the drivers "deeply" like X, pulseaudio, etc. seems like it'd be impossible to support. SELinux is just not going to work, probably dito. somewhat releated stuff like the audit interface / netlink (maybe that classifies as "deep" driver knowledge though).
Then there's the really crazy stuff where you have the same interfaces natively but they operate subtley differently in weird corner cases, DJB has a page on some stuff, but I doubt anyone sane enough to have commit privs. knows all of these (if anyone at all does).
But what about next week?
This is highly misleading way of interpreting that data. Esp. given the caveats that lwn.net has to do just tracking the companies for linux kernel development. From what I can tell they unpacked all packages in Debian, and then assigned the "corporate contributor" ... so all of OpenOffice and user space NFS goes to Sun, even though they did little more than a code dump, 10 years ago, in the later case.
In general I'd hope that "the community" is not using Solaris now because they see through BS stats. like the above, and know that Sun are much less of a participant in the community than Red Hat. But then, given the free ride Canonical gets sometimes, I do worry that it is just a case of network effects working against Sun.
Or not. The quote says "we are creeping towards $30 million" and that this would make Canonical "self-sustaining", for regular updates, which is their current model of very little development and a lot of marketing.
So, yeh, it's possible that "creeping towards" does actually mean "really close to, and getting closer every day" and not just "going from 1 to 2 million this year". But I bet against that, personally.
And who is paying that $30 mil, is it desktop people ... or self supporting server users like the French govt. / Wikipedia. My guess is that any money is coming from the later.
So at the end of the creep towards $30 mil, what do you get? Still not enough money to fund actual development (feel free to log bugs and RFEs at bugzilla.redhat.com), and a focus on servers instead of the desktop for what little work they do (because that's who is paying).
Then you probably want to read: Patrick Logan on why SMT isn't "awesomez".