The OS is more reliable then Windows, however the hardware can still fail. They use standard disks
Yes and no. They mount the disc in shock-absorbent rubber (which Toshiba/Compal didn't for my x86 laptop, relying on the rigidity of the plastic case to protect it) and there is a "sudden motion sensor" either in the laptop or the drive, which, under the control of software, can unload the heads before the laptop hits the ground. More details here.IBM do something similar on their Thinkpads.
So if you are not writing code of OSS then you are not entitled to an opinion? Also, she does contribute to things like look/feel/UI design etc.
Almost; I'd say there are about three scenarios where J. Random User's opinion counts for anything:
They're also a developer and submit a suitably tasteful patch (and maintain it until it is accepted into the upstream package) to add the desired functionality.
They're a customer of a vendor (this could include Debian, of course, in which case no money need change hands). By petitioning the vendor, the vendor can develop the feature or lobby developers on their behalf if the user's request is in alignment with the vendor's goals.
Depending on the developer, a well thought-out description (i.e. including a description of how to handle odd corner cases and any compatibility issues with the existing implementation, at the very least) may be enough to convince an altruistic developer to implement it.
I find the first of these options works best for me.
I've used Linux's built-in software RAID to access RAID0 and RAID1 arrays that were built from two USB-attached PATA drives (i.e. transplanted into USB caddies). The bottleneck was the USB 2.0 bus (performance was about half what I got when the same drives were previously attached to an on-motherboard Promise PDC20276 controller), but otherwise worked completely as expected. Further, Intel USB 2.0 controllers were about 10-15MB/s faster than NEC USB 2.0 controllers.
$240/year is a ton of money to be paying for just 1 network.
Not really; Murdoch's (aka 'Fox') basic 'Sky' package is 20GBP per month (in addition to the license, obviously:-), rising to about 40GBP per month with all the channels.
How much does the TV Licencing tax actually get you? How many channels does the BBC offer?
If you're still using an analogue terrestrial receiver, then you get two BBC channels, two privately-held channels funded by advertising (ITV + five), and Channel 4 (publicly owned, but funded entirely by advertising and other commercial activities).
If you're using a digital terrestrial receiver (20-30GBP one-off payment for the set-top box), then you can usually get all the Freeview channels - the two full BBC channels, BBC3/CBBC (new comedy and kids programmes, respectively - they share a single MUX), BBC4/CBeebies (documentaries, current affairs, arts and kids, respectively), abc1 (US shows), BBC, ITN and Sky News, Sky Sports News, Sky Travel, UKTV History, three music/yoof channels (TMF, The Hits, FTN), BBC Parliament, and a bunch of home shopping channels. You also get a whole raft of radio stations, including the BBC's eleven stations.
So, all in all, it's probably fairly comparable with your arrangements.
The whole TV Licencing thing is absurd anyhow. I never knew it existed before, but $150/year to use equipment that you have already paid for is moronic.
It's actually £121, so that's about US$240 at current rates.
I guess it pays for the BBC? Why don't they just go private and get advertising funds like everyone else?
Because then the BBC would produce (more) populist crap, just like the other advertising-funded channels do to get their ratings (and corresponding advertising sales revenue).
Oh, other disinformation I have heard on usenet is that if on their first visit they find you *do* have a TV setup for reception, then they give you the chance to pay your licence. Nope. Friend of mine let them in, they saw the TV set working, and they hit them with the £1000 fine or court case thing. Hence, thats why I wouldn't let them past the door!
That's not complete disinformation, but they might well be pursuing non-payers more aggressively now. I say this because when I was a student, back around 1993 or 1994, my shared house had a call from the licensing agency. Our TV was on and visible from outside the front door, so we were busted, but the official gave us a week or two to get a license. That might well have been an unofficial and unsanctioned action, but officials are human beings too and respond to both aggression and politeness the same as you or I.
What if I only use a TV to watch videos/DVDs/as a monitor for my games console? Do I still need a licence?
You need to notify us in writing that this is the case and one our Enforcement Officers may need to visit you to confirm that you do not need a licence.[My emphasis]
Isn't that almost exactly the experience I described in my first post on this matter? (TBH, I don't know whether said friend wrote in advance, or let them just find out on a routine check).
Also, nowhere have I asserted that deviating from 'normal' behaviour (i.e. owning a TV and having a license for it) doesn't attract a bit more attention/harassment than one might like. That said, some of the complaint letters linked from the marmalade.net site are somewhat inflammatory. If I was a call-centre worker on minimum wage and received one, I'd be tempted to put the sender on the shitlist just for the sake of it.
Meaning of "television receiver" 9. - (1) In Part 4 of the Act (licensing of TV reception), "television receiver" means any apparatus installed or used for the purpose of receiving (whether by means of wireless telegraphy or otherwise) any television programme service, whether or not it is installed or used for any other purpose.
His television isn't installed (i.e. tuned, or connected to an aerial) or "used for the purpose of receiving any television programme service" so it's not a television receiver in the eyes of the law.
No, it's legit because, as this post pointed out, the tvlicensing.co.uk website itself says that you only need a license if the equipment is being used to 'receive or record TV programmes'. His TV isn't, and, more importantly, the official was satisfied that this is the case.
Are you saying that you're routinely visited by guys to check if you have a TV license?!?
No, they routinely visit properties that don't have a license registered at that address. According to this site and USENET, though, you're not obliged to let them in unless they have a warrant.
We're not quite a police state yet, despite the previous and present government's moves to turn it into one (CJA, RIPA, ID Cards + ID database, Anti-Terrorism Act etc. etc.).
If you have a TV, but just use it attached to a dvd player without a licence, you are breaking the law.
No, you are wrong. A friend has a TV which he only uses as a display for retro consoles and home computers. A man from TV Licensing dropped by unannounced one day, observed that the aerial lead was disconnected and all channels detuned from those frequencies in use in his region, and declared that no license fee was payable.
But, why the hell does OpenOffice launch so slowly? I mean, really. Gnumeric launches in two seconds. Abiword launches in 4 seconds. OpenOffice (writer or calc) launches in 25 seconds?? An app that starts piggy, feels piggy.
Because it's a large application (it includes lots of cross-platform portability toolkit functions), written in C++. This affects Mozilla and KDE too. More info here.
If you're not using a distro that includes prelinking, you should upgrade. If you are, you should make sure that the prelink process runs regularly.
Perhaps the newer XEONs wih the huge on chip caches and faster FSBs fair better. I did another google, and every benchmark I find is rather old. I did find one newer one, but it only does a tiny set of tests on *one* platform with *one* product. THe newer benchmark does show the Intel doing better in some areas and the opteron doing better in others, but it's hard to compare it to older benchmarks because the old ones don't do these tests.
Here is the 'new' benchmark I found:
http://www.anandtech.com/IT/showdoc.aspx?i=2347
The anandtech ones have been the most useful comparisions that I've found. That said, benchmarks are only really a pointer; the only true test is to use your code as the benchmark. Unfortunately, I don't have the budget to buy one of each. I would love to spend a day or two benchmarking an Opteron system against a comparable Xeon system.
Your issue appears to be that you will ony buy from a select few vendors.
Yeah, that's a bit of it; potentially any vendor is possible, providing our preferred vendors don't sell what we want (e.g. our preferred x86 supplier hasn't launched their full range of EM64T Xeons yet, so we're going to buy from Dell instead). But there's the 'no-one ever got fired for buying X' situation; if your recommendation doesn't work out...
Without giving pointers to the benchmarks you've seen, continuing this discussion is somewhat pointless.
Something I will note is that Intel have responded to competition from the Opteron and significantly revised the design of the Xeon several times now - upping the memory bus from 400MHz to 533MHz to 800MHz with the current models. Further, the amount of cache varies dramatically too; from 256KB L2+512KB L3 to 512KB L2+4MB L3 to 2MB L2 + 0MB L3 - this table gives you the rundown. Saying 'Opteron beats Xeon by x%' isn't really very informative unless you specify which model of Xeon you're talking about.
Finally, unless I'm wearing my computer systems architecture hat (I still have a soft spot for it, as it was my official degree title:), I don't care about performance/clock cycle, only about performance/£. Systems based around AMD's chips generally thrash Intel's on the former, whilst not always being so convincing on the latter (usually due to higher chipset/motherboard prices, IME).
In terms of SMP performance, Opteron systems are about 80% faster that XEON systems. If you are going to run some sort of database on your server, an Opteron solution that is 10-20% more expensive would be worth it.
That would indeed be sound reasoning, if that was reflected by the benchmarks I saw.
Have you even seen the reviews/benchmarks that compare the SMP XEON to SMP Opteron systems?
The ones on Anandtech (find by doing an 'advanced' search for 'opteron xeon' as 'all these words')?
If so, yes, I did. The benchmarks I saw (Sep 13 2004) led me to infer that a 2xXeon 3.2 1MB system would be at most 15% slower than a system with 2xOpteron 2.4s. Further, in order to get prices that were no more than 10% higher than the comparable Xeon system, we'd need to buy from a non-preferred vendor. Sun were far worse due to the need to go with a v40z for >1 133MHz PCI-X slots in one machine and >2 discs in the other, and that was with our discount (admittedly, the v40z is expandable to 4-way).:(
I'd be very interested in seeing better benchmarks for future reference, particularly if they use x86_64 Linux, Apache, MySQL/PostgreSQL, etc as the benchmark platform, rather than Windows.
Price conscience IT departments wont like the price for a SMP Xeon server compared to an SMP Opteron.
Actually, as someone who's just priced up a couple of specifications for beefy x86 machines recently, the Xeon-based system came out somewhat cheaper; the systems were as similar as I could get (one with 4GB RAM, a single ATA disc, the other with 8GB RAM and 4x143GB SCSI discs RAIDed). The Xeon-based systems were both 2xXeon 3.2 Nocona (1MB L2), the Opteron-based systems were both 2xOpteron 250 2.4GHz. I don't have the quotes in front of me, but I seem to recall the Opteron systems were about 10-20% more expensive.
Further, of our preferred vendors, only Sun sold Opteron based machines, and their Opteron systems are significantly more expensive to us than the same Newisys hardware bought from other vendors.
Finally, all but Sun's v40z/Newisys 4300 were limited in terms of 133MHz PCI-X slots. This is important for the role for which the 4GB machine is to be used.
I remain open to Opterons, and I really like the idea of each CPU having its own memory bus. But now that AMD are doing all the same sorts of things that Intel do with their platform (as opposed to merely CPU offerings), you can't assume that the Opteron systems will be cheaper, or even better value than similar Xeon systems.
i dont mean to troll, but i seriously dont understand how anybody can use a distro without package management tools as good as aptget or portage.
I agree entirely, which is why I invested the time in learning how to use rpm properly - it can do everything that portage and dpkg can do (note that rpm is not and was never meant to be analogous to apt-get; yum or the rpm port of apt are the appropriate comparisons).
Of course, what makes Gentoo and Debian special is the number of packages that are available as native distro packages from the respective standard repositories. RH/Fedora has been behind both those distros for a long time, but with Fedora and repositories such as freshrpms.net, dag and atrpms, it's catching up.
If a ready-built package isn't available for the version of the package you want to install, then it does come down to rolling your own RPMs if you want to keep your system sane. This is the same situation as exists for Gentoo and Debian, but most users don't experience this as a) they're not distro developers or package maintainers and b) they find new enough versions of the packages they wish to use already present in the standard repositories (because some distro developer has already done the work for them already).
I've read the instructions for creating your own Gentoo Ebuilds, and the process is almost identical in complexity (or simplicity, if you prefer) to creating your own RPMs. Debian is a bit easier, because it's considered acceptable to roll all the changes you make into a single über-patch.
In the end, I've stuck with RH/Fedora as I want to continue to leverage the experience I've built up from running it over the last ten years. I also usually agree with their package/version choices and the changes they make to the upstream packages.
I've already mentioned Centos and Whitebox, and that's what I meant by 'and friends'.
So instead of using RH you suggest I go to a third party leech?
No, I'm suggesting you take advantage of the precise freedoms that using Free software gives you (if you feel that's your best option). Red Hat must even tacitly approve of CentOS and Whitebox, or they would only distribute source to people who bought RHEL subscriptions (and then only tarballs + patches, rather than the complete src.rpms that they do currently) as allowed by the GPL.
A process which still won't get reasonably priced paid support?
Commercial Support for CentOS. Whether those supporters provide good value compared with DIY, RH, or SuSE is subjective and up to you to decide. Once again, this is another freedom provided by using Free software. Also, bear in mind that the prices quoted on RH's site aren't necessarily the complete story. If you're buying for an enterprise, you'll probably be able to get a deal out of them, as you would when buying support or software from any other enterprise vendor.
Essentially, your argument boils down to you being upset that if you want RH support, their list price indicates you need to pay an extra US$79 for RHEL WS compared with RHL back in 2002 or so. Services generally go up in price over time, deal with it. It's called inflation.
And as a leech as opposed to an actual customer I'd have even less leverage in getting broken stuff fixed, wouldn't I?
Well, yes. But that's exactly the same situation you would have been in prior to RHEL/Fedora if you were using cheap*bytes or Linux Emporium CDs. If you want support, pay for it from someone who you believe offers good VFM. If you don't pay, it's pointless bitching that you don't get it.
Personally, I'm OK with a distro upgrade every year or so on my workstations
Well good for you. Many people find they have more important things to do (actually productive work) than upgrade workstations.
Well, its either that or lose time to rebuilding massive chunks of the OS stack from source in order to get new applications (aka my productive work) running on a year-old distro. I'm lazy, so I prefer to do an upgrade.:-)
Slashdot Mirror
I'm not sure whether you meant this as wit, but laptops with RAID0 and RAID1 are now available. I fear for the battery life and heat output...
The OS is more reliable then Windows, however the hardware can still fail. They use standard disks
Yes and no. They mount the disc in shock-absorbent rubber (which Toshiba/Compal didn't for my x86 laptop, relying on the rigidity of the plastic case to protect it) and there is a "sudden motion sensor" either in the laptop or the drive, which, under the control of software, can unload the heads before the laptop hits the ground. More details here. IBM do something similar on their Thinkpads.
Almost; I'd say there are about three scenarios where J. Random User's opinion counts for anything:
They're also a developer and submit a suitably tasteful patch (and maintain it until it is accepted into the upstream package) to add the desired functionality.
They're a customer of a vendor (this could include Debian, of course, in which case no money need change hands). By petitioning the vendor, the vendor can develop the feature or lobby developers on their behalf if the user's request is in alignment with the vendor's goals.
Depending on the developer, a well thought-out description (i.e. including a description of how to handle odd corner cases and any compatibility issues with the existing implementation, at the very least) may be enough to convince an altruistic developer to implement it.
I find the first of these options works best for me.
This is, as far as I know, what the larger USB discs (e.g. Lacie) do; a RAID0 array between two (or more) smaller discs.
I've used Linux's built-in software RAID to access RAID0 and RAID1 arrays that were built from two USB-attached PATA drives (i.e. transplanted into USB caddies). The bottleneck was the USB 2.0 bus (performance was about half what I got when the same drives were previously attached to an on-motherboard Promise PDC20276 controller), but otherwise worked completely as expected. Further, Intel USB 2.0 controllers were about 10-15MB/s faster than NEC USB 2.0 controllers.
Lots of hardware RAID controllers use the Intel i960 as the embedded controller. I've read that 66MHz models are only slightly faster than a PII-400.
Not really; Murdoch's (aka 'Fox') basic 'Sky' package is 20GBP per month (in addition to the license, obviously :-), rising to about 40GBP per month with all the channels.
How much does the TV Licencing tax actually get you? How many channels does the BBC offer?
If you're still using an analogue terrestrial receiver, then you get two BBC channels, two privately-held channels funded by advertising (ITV + five), and Channel 4 (publicly owned, but funded entirely by advertising and other commercial activities).
If you're using a digital terrestrial receiver (20-30GBP one-off payment for the set-top box), then you can usually get all the Freeview channels - the two full BBC channels, BBC3/CBBC (new comedy and kids programmes, respectively - they share a single MUX), BBC4/CBeebies (documentaries, current affairs, arts and kids, respectively), abc1 (US shows), BBC, ITN and Sky News, Sky Sports News, Sky Travel, UKTV History, three music/yoof channels (TMF, The Hits, FTN), BBC Parliament, and a bunch of home shopping channels. You also get a whole raft of radio stations, including the BBC's eleven stations.
So, all in all, it's probably fairly comparable with your arrangements.
It's actually £121, so that's about US$240 at current rates.
I guess it pays for the BBC? Why don't they just go private and get advertising funds like everyone else?
Because then the BBC would produce (more) populist crap, just like the other advertising-funded channels do to get their ratings (and corresponding advertising sales revenue).
That's not complete disinformation, but they might well be pursuing non-payers more aggressively now. I say this because when I was a student, back around 1993 or 1994, my shared house had a call from the licensing agency. Our TV was on and visible from outside the front door, so we were busted, but the official gave us a week or two to get a license. That might well have been an unofficial and unsanctioned action, but officials are human beings too and respond to both aggression and politeness the same as you or I.
What if I only use a TV to watch videos/DVDs/as a monitor for my games console? Do I still need a licence?
You need to notify us in writing that this is the case and one our Enforcement Officers may need to visit you to confirm that you do not need a licence. [My emphasis]
Isn't that almost exactly the experience I described in my first post on this matter? (TBH, I don't know whether said friend wrote in advance, or let them just find out on a routine check).
Also, nowhere have I asserted that deviating from 'normal' behaviour (i.e. owning a TV and having a license for it) doesn't attract a bit more attention/harassment than one might like. That said, some of the complaint letters linked from the marmalade.net site are somewhat inflammatory. If I was a call-centre worker on minimum wage and received one, I'd be tempted to put the sender on the shitlist just for the sake of it.
His television isn't installed (i.e. tuned, or connected to an aerial) or "used for the purpose of receiving any television programme service" so it's not a television receiver in the eyes of the law.
No, it's legit because, as this post pointed out, the tvlicensing.co.uk website itself says that you only need a license if the equipment is being used to 'receive or record TV programmes'. His TV isn't, and, more importantly, the official was satisfied that this is the case.
No, they routinely visit properties that don't have a license registered at that address. According to this site and USENET, though, you're not obliged to let them in unless they have a warrant.
We're not quite a police state yet, despite the previous and present government's moves to turn it into one (CJA, RIPA, ID Cards + ID database, Anti-Terrorism Act etc. etc.).
They already are, and they run Linux.
No, you are wrong. A friend has a TV which he only uses as a display for retro consoles and home computers. A man from TV Licensing dropped by unannounced one day, observed that the aerial lead was disconnected and all channels detuned from those frequencies in use in his region, and declared that no license fee was payable.
Because it's a large application (it includes lots of cross-platform portability toolkit functions), written in C++. This affects Mozilla and KDE too. More info here.
If you're not using a distro that includes prelinking, you should upgrade. If you are, you should make sure that the prelink process runs regularly.
S'ok! ;-)
Perhaps the newer XEONs wih the huge on chip caches and faster FSBs fair better. I did another google, and every benchmark I find is rather old. I did find one newer one, but it only does a tiny set of tests on *one* platform with *one* product. THe newer benchmark does show the Intel doing better in some areas and the opteron doing better in others, but it's hard to compare it to older benchmarks because the old ones don't do these tests. Here is the 'new' benchmark I found: http://www.anandtech.com/IT/showdoc.aspx?i=2347
The anandtech ones have been the most useful comparisions that I've found. That said, benchmarks are only really a pointer; the only true test is to use your code as the benchmark. Unfortunately, I don't have the budget to buy one of each. I would love to spend a day or two benchmarking an Opteron system against a comparable Xeon system.
Anyhow - good luck to you.
Cheers!
Right, that benchmark tests the old 533MHz FSB Xeons with the old E7501 chipset.
Yeah, that's a bit of it; potentially any vendor is possible, providing our preferred vendors don't sell what we want (e.g. our preferred x86 supplier hasn't launched their full range of EM64T Xeons yet, so we're going to buy from Dell instead). But there's the 'no-one ever got fired for buying X' situation; if your recommendation doesn't work out...
Something I will note is that Intel have responded to competition from the Opteron and significantly revised the design of the Xeon several times now - upping the memory bus from 400MHz to 533MHz to 800MHz with the current models. Further, the amount of cache varies dramatically too; from 256KB L2+512KB L3 to 512KB L2+4MB L3 to 2MB L2 + 0MB L3 - this table gives you the rundown. Saying 'Opteron beats Xeon by x%' isn't really very informative unless you specify which model of Xeon you're talking about.
Finally, unless I'm wearing my computer systems architecture hat (I still have a soft spot for it, as it was my official degree title :), I don't care about performance/clock cycle, only about performance/£. Systems based around AMD's chips generally thrash Intel's on the former, whilst not always being so convincing on the latter (usually due to higher chipset/motherboard prices, IME).
That would indeed be sound reasoning, if that was reflected by the benchmarks I saw.
Have you even seen the reviews/benchmarks that compare the SMP XEON to SMP Opteron systems?
The ones on Anandtech (find by doing an 'advanced' search for 'opteron xeon' as 'all these words')?
If so, yes, I did. The benchmarks I saw (Sep 13 2004) led me to infer that a 2xXeon 3.2 1MB system would be at most 15% slower than a system with 2xOpteron 2.4s. Further, in order to get prices that were no more than 10% higher than the comparable Xeon system, we'd need to buy from a non-preferred vendor. Sun were far worse due to the need to go with a v40z for >1 133MHz PCI-X slots in one machine and >2 discs in the other, and that was with our discount (admittedly, the v40z is expandable to 4-way). :(
I'd be very interested in seeing better benchmarks for future reference, particularly if they use x86_64 Linux, Apache, MySQL/PostgreSQL, etc as the benchmark platform, rather than Windows.
Actually, as someone who's just priced up a couple of specifications for beefy x86 machines recently, the Xeon-based system came out somewhat cheaper; the systems were as similar as I could get (one with 4GB RAM, a single ATA disc, the other with 8GB RAM and 4x143GB SCSI discs RAIDed). The Xeon-based systems were both 2xXeon 3.2 Nocona (1MB L2), the Opteron-based systems were both 2xOpteron 250 2.4GHz. I don't have the quotes in front of me, but I seem to recall the Opteron systems were about 10-20% more expensive.
Further, of our preferred vendors, only Sun sold Opteron based machines, and their Opteron systems are significantly more expensive to us than the same Newisys hardware bought from other vendors.
Finally, all but Sun's v40z/Newisys 4300 were limited in terms of 133MHz PCI-X slots. This is important for the role for which the 4GB machine is to be used.
I remain open to Opterons, and I really like the idea of each CPU having its own memory bus. But now that AMD are doing all the same sorts of things that Intel do with their platform (as opposed to merely CPU offerings), you can't assume that the Opteron systems will be cheaper, or even better value than similar Xeon systems.
I agree entirely, which is why I invested the time in learning how to use rpm properly - it can do everything that portage and dpkg can do (note that rpm is not and was never meant to be analogous to apt-get; yum or the rpm port of apt are the appropriate comparisons).
Of course, what makes Gentoo and Debian special is the number of packages that are available as native distro packages from the respective standard repositories. RH/Fedora has been behind both those distros for a long time, but with Fedora and repositories such as freshrpms.net, dag and atrpms, it's catching up.
If a ready-built package isn't available for the version of the package you want to install, then it does come down to rolling your own RPMs if you want to keep your system sane. This is the same situation as exists for Gentoo and Debian, but most users don't experience this as a) they're not distro developers or package maintainers and b) they find new enough versions of the packages they wish to use already present in the standard repositories (because some distro developer has already done the work for them already).
I've read the instructions for creating your own Gentoo Ebuilds, and the process is almost identical in complexity (or simplicity, if you prefer) to creating your own RPMs. Debian is a bit easier, because it's considered acceptable to roll all the changes you make into a single über-patch.
In the end, I've stuck with RH/Fedora as I want to continue to leverage the experience I've built up from running it over the last ten years. I also usually agree with their package/version choices and the changes they make to the upstream packages.
then they should be registered with the Charity Commission, whose register of charities is searchable.
So instead of using RH you suggest I go to a third party leech?
No, I'm suggesting you take advantage of the precise freedoms that using Free software gives you (if you feel that's your best option). Red Hat must even tacitly approve of CentOS and Whitebox, or they would only distribute source to people who bought RHEL subscriptions (and then only tarballs + patches, rather than the complete src.rpms that they do currently) as allowed by the GPL.
A process which still won't get reasonably priced paid support?
Commercial Support for CentOS. Whether those supporters provide good value compared with DIY, RH, or SuSE is subjective and up to you to decide. Once again, this is another freedom provided by using Free software. Also, bear in mind that the prices quoted on RH's site aren't necessarily the complete story. If you're buying for an enterprise, you'll probably be able to get a deal out of them, as you would when buying support or software from any other enterprise vendor.
Essentially, your argument boils down to you being upset that if you want RH support, their list price indicates you need to pay an extra US$79 for RHEL WS compared with RHL back in 2002 or so. Services generally go up in price over time, deal with it. It's called inflation.
And as a leech as opposed to an actual customer I'd have even less leverage in getting broken stuff fixed, wouldn't I?
Well, yes. But that's exactly the same situation you would have been in prior to RHEL/Fedora if you were using cheap*bytes or Linux Emporium CDs. If you want support, pay for it from someone who you believe offers good VFM. If you don't pay, it's pointless bitching that you don't get it.
Personally, I'm OK with a distro upgrade every year or so on my workstations
Well good for you. Many people find they have more important things to do (actually productive work) than upgrade workstations.
Well, its either that or lose time to rebuilding massive chunks of the OS stack from source in order to get new applications (aka my productive work) running on a year-old distro. I'm lazy, so I prefer to do an upgrade. :-)