Maplin still sell components (and in the small quantities a hobbyist needs - unlike RS), but yes, they've also been selling "toys for boys" and general electronic tat for a few years now. I look upon it as getting the masses to subsidise a high-street electronic components store for us geeks, so I'm happy to tolerate it as long as they don't start cutting down their component list.
I'm opposed to the per-tv "License Fee" charged by the British government on philosphical grounds.
Extracting a tax for simply owning a television set...
As has been pointed out in other threads, the licence is for as many TVs as you like in a single household, NOT per TV.
...creates a captive audience and the quality of the programming suffers as a result. For every Monty Python's Flying Circus there are countless shows that wouldn't make it past the pilot phase here in the states. If the quality of programming on BBC-america is any indication, brits are being robbed.
You should see the (lack of) quality on our major commercial analogue channels then; one wants to be a 24x7 soap/Hitler/pr0n/US TV Movies channel, and the other is dumbed-down sensationalist news/soap/reality TV/blockbuster hollywood movies.
Only Channel 4 has any quality programmes that compare with the BBC's, and lots of those are being bought in from the likes of the Discovery and History channels these days (it's easy to tell them from the programmes they make themselves as the bought-in programmes repeat themselves every 5-10 minutes for where the ad. breaks would be).
I'm curious about something, and maybe some of you british slashdotters can answer some questions for me. What else is there on the air other than the BBC?
BBC1: soap, blockbuster films, major sporting/cultural events, sport, investigative journalism, news, popular comedy, minority sports (e.g. darts, snooker). Funded by licence fee.
BBC2: documentaries, arts, investigative journalism, economics/finance, science, history, art/cult films, new comedy. Funded by licence fee.
ITV: soap, reality TV, sport, blockbuster films, sport, sensationalist news, regional content. Privately owned and funded exclusively by advertising.
Channel 4:much like BBC2, but more mainstream content (e.g. some soaps). Minority sports include horse racing and various non-European sports. Publically owned, but funded exclusively by advertising.
Channel 5:soap, Hitler documentaries, softcore pr0n, old blockbusters, US TV movies. Privately owned, funded by advertising (and deep pockets, since the last I heard, it wasn't doing very well).
With Freeview (free-of-subscription charge digital TV), you get BBC News 24, The History Channel, Parliament, CBeebies (kids TV) and some ITV and BBC channels mostly used for repeats right now. Oh, and the usual set of home shopping channels and suchlike.
How many channels do brits generally have to choose from?
I would guess most AB social class homes only have the five broadcast analogue channels I described earlier, or maybe Freeview.
Is cable TV common, and if so what kind of channels are there on it?
Other social classes are more likely to have Murdoch-owned SkyTV with programming most North Americans would be familiar with, I'm sure - Fox, SciFi, Nick, etc.
Do you get HBO?
Channel 4 and, especially, Channel 5 show quite a lot of HBO-sourced material. I think HBO is available from Sky.
ESPN?
Sports? Dunno. A lot of the content wouldn't be of much interest in the UK. Soccer has a much bigger following here than American Football, Basketball, etc. and SkySports + BBC + ITV cover that well.
I was told when I was a little kid that the BBC was the ONLY channel available over there. I find that hard to believe. Imagine if the only channel americans had to watch was PBS.
No, that's absolutely correct. We don't have running water, sanitation, or electricity either. Also, we all have tea with the Queen each week, after kissing her feet.;-)
yum and apt are not replacements for rpm, they're just frontends for rpm that handle automagic dependency resolution.
Take it from me; as long as you stick with sensibly built packages from trustworthy repositories (e.g. RH, Freshrpms), your RPM database will stay sane.
I've done RH6.2 directly to RH7.3. I was wary, but it worked just fine, bar our own applications that needed rebuilding against the new versions of the libraries included in RH7.3 (which is entirely expected).
Were you replacing random bits of the distro with built-from-source packages, by any chance?
One key used for signing 10's of millions of IDs, issued by multiple entities is virtually guarenteed to "leak". Even if multiple keys are used they would still need to be secure for the entire life of the IDs they are used to sign.
The solution (proposed by someone else on the ukcrypto list) is to sign hashes generated with multiple algorithms with multiple keys. That way, if a key is compromised, or a hash algorithm is subsequently found to be less secure than previously thought, there is no need to (expensively!) invalidate the entire installed base of ID cards as there are other signatures that can still be "relied" upon.
All that proves is that the card and the cardholder match. Putting fingerprints, retina scan or whatever information on the card dosn't address this issue any more than having a photograph (the most common form of current biometrics) on the card.
I'm wholeheartedly opposed to "infallible" biometric/smart ID cards, but yours is a flawed argument. The biometric information could be digitally signed using a government key at the point of issue.
Of course, this doesn't eliminate the possibility of corrupt issuing staff, a compromised key, or other, similar, attacks. If that was your point, you should have made it clearer.:-)
Not yet; it turned out I had enough olde fashioned Pb/Sn solder for the repair I bought it for.
My soldering iron does go to 300 or 400 deg. C, though, so I might be OK. That said, I've heard that the switch to non-Pb solder industry-wide has resulted in a higher rate of dry joints (and other soldering defects) which in turn results in a higher rate of failure, which in turn results in more landfill.:(
Unanticipated consequential effects are a bitch, eh?
...and besides, it would not be the "open-source community" that is having difficulties delivering - it would be SuSE. The buck stops with them and any of their partners who signed the contract.
Then you'd be amazed at how often kernel developers use this argument.
"Oh, you're running the NVIDIA module, that taints the kernel, so we won't even listen to your complaints about stability, because that simply has to be the cause."
I think you're overstating what the kernel developers are saying. A more accurate representation would be:
"Oh, you're running the NVIDIA module, that taints the kernel, so we won't even listen to your complaints about stability, because it could well be the cause. Because we don't have source for it, we have no way of telling for certain if it is the cause or not. If it is the cause, then it would be a waste of my time to investigate your problem, when I've got Free code I could be hacking on and improving. Check with NVIDIA first, if they say the problem's definitely not in their code and explain why, come back to me."
How many Dell, HP, Compaq, graybox etc.... hardware rollouts are greeted with the same kind of fervor that Apple computer hardware announcements inspire?
Actually, the same thing happens with PeeCee/commodity hardware too; try and find a new 10GByte disc for 10GBP, or a new Celeron 500 for 5GBP. This is particularly annoying if you want to put together some ultra-cheap new machines (for an undemanding user such as Aunt Tilly, say) without resorting to the skankiest hardware out there (which is probably still more expensive than obsolete stuff would be, if it were still available).
A couple of members of my LUG are big Mac fans, and they've done a fine job of showing us how well modern Macs work - traditional Apple attention-to-detail and UI niceness combined with the power and stability of UNIX; what more could a user want?
Even price isn't a major stumbling block these days; myself and one of the Mac guys compared the laptops we'd bought about a year previously; the Mac was equal or better in every respect, looked prettier, and only cost about 100GBP more (which IMHO, is pretty good VFM considering Apple's customer service, and how well the OS and the hardware work together compared with PC hardware and either Windows or Linux).
Unfortunately, there are a couple of showstoppers for me. These are:
Availability of Free source code for the entire OS stack (i.e. not just Darwin). Sorry, but even though I don't use the source code for some OS components at all at the moment, I've become hooked on havinng it available - a security blanket if you like.
Availability of Mac hardware from multiple vendors. I'm not keen on being tied to a single hardware vendor. Of course, if this was the case, chances are we'd begin to see the same hardware/OS incompatibilities the PC world is plagued with.:(
Rapidly diminishing: availability of some applications only as x86 binaries (e.g. closed multimedia CoDecs). The status of OpenOffice.org/MacOS X was a showstopper when I was buying my laptop, but this is no longer a problem.
This is why we have version numbers on.so files. If your program needs libfoo.so.2.1, than it should list that as a dependency, rather than just libfoo.so, which might actually be a link to libfoo.so.1.0 (without some bugfix or feature).
Unfortunately, there have been many cases of libraries changing their APIs without changing the major version number (e.g. glibc). So you can't rely on the library major version number as you would like to do.
As far as I can tell, the argument goes that if you change the major version too frequently, you'll end up with lots of versions of the library present in memory, depending on which version a binary was compiled against.
I'm not saying that I go along entirely with that line of reasoning, but I present it for reasons of balance.
Personally, I think AmigaOS' implementation of shared libraries were about the tidiest I've seen.
Incidentally, RH/Fedora has started adopting a number of Debian-isms; both MySQL and PostgreSQL provide the 'database' virtual package (so either satisfy an application's dependency for 'database') and Sendmail and Postfix can be switched using a Debian-like alternatives scheme.
The funny thing is, last I used RPM (I am on Debian now, which obviates the need for looking back), it actually reported the _files_ a package depended on. How hard could it be, then, to check for the existence of these files (rather than whether a package that says it provides them is installed)? How the files got there doesn't matter, if the dependencies are met, the installed program works.
Except, if those files (or their sources) required patching in order to make a dependent program work, then having a random version of those files won't make that program work - though it may look that way until you come to test it (thoroughly). This is partly why package dependencies are preferred to letting RPM figure out file dependencies automagically. It also helps the administrator in figuring out which packages he needs to install (because RPM will report that package names are required, rather than file names).
Anyway, talk of "RPM dependency hell" is somewhat redundant now; with apt-get (yes, we know Debian had it first) and yum and, most importantly, robustly-buit packages (e.g. Red Hat's, and freshrpms.net), dependencies are resolved automatically.
Why would anyone need to pull in libMD5 as a dependancy, to get what amounts to three functions in under 100 lines of code?
From bitter experience, and from the last seven years working in computer security, I've learnt that for most things, there's at least one person in the world writing and releasing open software who understands the problem-space better than I. Rather than trying to reinvent the wheel, and learning all the snags and gotchas the hard way, I use their mature, battle-hardened code and get on with life. Of course, that doesn't prevent me checking return codes and using assert() regularly, as all paranoid programmers should.
Based on your comments, either you're in the top 0.1% of software authors out there, or you have an over-inflated sense of your own abilities.:-)
I tend to agree, but I have found one case on Redhat where RPMs give me nothing but trouble: Perl.
I have always had problems with Perl when I go to install a new module from CPAN if Perl was installed with an RPM file or came with the system (i.e. installed when the system was installed). Perl itself works great, but some CPAN packages barf when I try to compile them.
...import duty (2%, AFAICS) and VAT (17.5%), unless you're planning on smuggling the computer in. Also remember that your warranty may well be non-existant. By the time you've added ~20% for duty and VAT, you may find the savings don't make privately importing worthwhile.
I make it your saving is under 300GBP, without taking account of the warranty, plane ticket/lost time from holiday or mailbox rental.
All down to the cost of labour and the costs of running a business, I'm afraid.
I don't know what it's like in the US, but here in the UK, the cost of new PCs is making PC "repairs" uneconomic if the repairer wants to charge rates similar to those of plumbers and the like (to put some numbers on that, a typical rate for a plumber is 60GBP per hour, and a new PC costs from 300GBP, with monitor and preloaded copy of whatever the latest flavour of Windows is; how much work do you reckon can do in under 5 hours?)
Of course, this does discount the stupid and the penny-wise-pound-foolish, whom are probably the best cash cows out there for any business.
("TEST=" in packets traversing all versions of FW-1 unblocked up until around 2 years ago anyone?)
/me doubletakes.
Have you got any more information on that? That appears to be a serious vulnerability that I hadn't previously heard about... Securityfocus.com's vulndb doesn't seem to know about it, either.
(Okay, just for kicks, here's an actual tidbit of useful Checkpoint info: There's a Rule Zero. It doesn't appear in the rules screen. It's probably not doing what you think it's doing.)
I'm not a great fan of FW-1, but it's a shame that whoever plunked down the cash for the software didn't also pay for some training for you; that snippet would have been covered in the basic course.
Slashdot Mirror
--
--
Extracting a tax for simply owning a television set...
As has been pointed out in other threads, the licence is for as many TVs as you like in a single household, NOT per TV.
You should see the (lack of) quality on our major commercial analogue channels then; one wants to be a 24x7 soap/Hitler/pr0n/US TV Movies channel, and the other is dumbed-down sensationalist news/soap/reality TV/blockbuster hollywood movies.
Only Channel 4 has any quality programmes that compare with the BBC's, and lots of those are being bought in from the likes of the Discovery and History channels these days (it's easy to tell them from the programmes they make themselves as the bought-in programmes repeat themselves every 5-10 minutes for where the ad. breaks would be).
I'm curious about something, and maybe some of you british slashdotters can answer some questions for me. What else is there on the air other than the BBC?
BBC1: soap, blockbuster films, major sporting/cultural events, sport, investigative journalism, news, popular comedy, minority sports (e.g. darts, snooker). Funded by licence fee.
BBC2: documentaries, arts, investigative journalism, economics/finance, science, history, art/cult films, new comedy. Funded by licence fee.
ITV: soap, reality TV, sport, blockbuster films, sport, sensationalist news, regional content. Privately owned and funded exclusively by advertising.
Channel 4:much like BBC2, but more mainstream content (e.g. some soaps). Minority sports include horse racing and various non-European sports. Publically owned, but funded exclusively by advertising.
Channel 5:soap, Hitler documentaries, softcore pr0n, old blockbusters, US TV movies. Privately owned, funded by advertising (and deep pockets, since the last I heard, it wasn't doing very well).
With Freeview (free-of-subscription charge digital TV), you get BBC News 24, The History Channel, Parliament, CBeebies (kids TV) and some ITV and BBC channels mostly used for repeats right now. Oh, and the usual set of home shopping channels and suchlike.
How many channels do brits generally have to choose from?
I would guess most AB social class homes only have the five broadcast analogue channels I described earlier, or maybe Freeview.
Is cable TV common, and if so what kind of channels are there on it?
Other social classes are more likely to have Murdoch-owned SkyTV with programming most North Americans would be familiar with, I'm sure - Fox, SciFi, Nick, etc.
Do you get HBO?
Channel 4 and, especially, Channel 5 show quite a lot of HBO-sourced material. I think HBO is available from Sky.
ESPN?
Sports? Dunno. A lot of the content wouldn't be of much interest in the UK. Soccer has a much bigger following here than American Football, Basketball, etc. and SkySports + BBC + ITV cover that well.
I was told when I was a little kid that the BBC was the ONLY channel available over there. I find that hard to believe. Imagine if the only channel americans had to watch was PBS.
No, that's absolutely correct. We don't have running water, sanitation, or electricity either. Also, we all have tea with the Queen each week, after kissing her feet. ;-)
--
http://www.priceguideuk.com/
http://www.internet-buyer.co.uk/
http://www.kelkoo.co.uk/
http://uk.pricerunner.com/
Oh, and buy with a Barclaycard to get a 60 day price promise on purchases over 50GBP and an extra years' warranty on purchases over 150GBP.
--
--
Take it from me; as long as you stick with sensibly built packages from trustworthy repositories (e.g. RH, Freshrpms), your RPM database will stay sane.
--
Were you replacing random bits of the distro with built-from-source packages, by any chance?
--
The solution (proposed by someone else on the ukcrypto list) is to sign hashes generated with multiple algorithms with multiple keys. That way, if a key is compromised, or a hash algorithm is subsequently found to be less secure than previously thought, there is no need to (expensively!) invalidate the entire installed base of ID cards as there are other signatures that can still be "relied" upon.
--
I'm wholeheartedly opposed to "infallible" biometric/smart ID cards, but yours is a flawed argument. The biometric information could be digitally signed using a government key at the point of issue.
Of course, this doesn't eliminate the possibility of corrupt issuing staff, a compromised key, or other, similar, attacks. If that was your point, you should have made it clearer. :-)
--
My soldering iron does go to 300 or 400 deg. C, though, so I might be OK. That said, I've heard that the switch to non-Pb solder industry-wide has resulted in a higher rate of dry joints (and other soldering defects) which in turn results in a higher rate of failure, which in turn results in more landfill. :(
Unanticipated consequential effects are a bitch, eh?
--
--
--
"Oh, you're running the NVIDIA module, that taints the kernel, so we won't even listen to your complaints about stability, because that simply has to be the cause."
I think you're overstating what the kernel developers are saying. A more accurate representation would be:
"Oh, you're running the NVIDIA module, that taints the kernel, so we won't even listen to your complaints about stability, because it could well be the cause. Because we don't have source for it, we have no way of telling for certain if it is the cause or not. If it is the cause, then it would be a waste of my time to investigate your problem, when I've got Free code I could be hacking on and improving. Check with NVIDIA first, if they say the problem's definitely not in their code and explain why, come back to me."
--
Actually, the same thing happens with PeeCee/commodity hardware too; try and find a new 10GByte disc for 10GBP, or a new Celeron 500 for 5GBP. This is particularly annoying if you want to put together some ultra-cheap new machines (for an undemanding user such as Aunt Tilly, say) without resorting to the skankiest hardware out there (which is probably still more expensive than obsolete stuff would be, if it were still available).
--
Even price isn't a major stumbling block these days; myself and one of the Mac guys compared the laptops we'd bought about a year previously; the Mac was equal or better in every respect, looked prettier, and only cost about 100GBP more (which IMHO, is pretty good VFM considering Apple's customer service, and how well the OS and the hardware work together compared with PC hardware and either Windows or Linux).
Unfortunately, there are a couple of showstoppers for me. These are:
Availability of Free source code for the entire OS stack (i.e. not just Darwin). Sorry, but even though I don't use the source code for some OS components at all at the moment, I've become hooked on havinng it available - a security blanket if you like.
Availability of Mac hardware from multiple vendors. I'm not keen on being tied to a single hardware vendor. Of course, if this was the case, chances are we'd begin to see the same hardware/OS incompatibilities the PC world is plagued with. :(
Rapidly diminishing: availability of some applications only as x86 binaries (e.g. closed multimedia CoDecs). The status of OpenOffice.org/MacOS X was a showstopper when I was buying my laptop, but this is no longer a problem.
--
--
Unfortunately, there have been many cases of libraries changing their APIs without changing the major version number (e.g. glibc). So you can't rely on the library major version number as you would like to do.
As far as I can tell, the argument goes that if you change the major version too frequently, you'll end up with lots of versions of the library present in memory, depending on which version a binary was compiled against.
I'm not saying that I go along entirely with that line of reasoning, but I present it for reasons of balance.
Personally, I think AmigaOS' implementation of shared libraries were about the tidiest I've seen.
Incidentally, RH/Fedora has started adopting a number of Debian-isms; both MySQL and PostgreSQL provide the 'database' virtual package (so either satisfy an application's dependency for 'database') and Sendmail and Postfix can be switched using a Debian-like alternatives scheme.
--
Except, if those files (or their sources) required patching in order to make a dependent program work, then having a random version of those files won't make that program work - though it may look that way until you come to test it (thoroughly). This is partly why package dependencies are preferred to letting RPM figure out file dependencies automagically. It also helps the administrator in figuring out which packages he needs to install (because RPM will report that package names are required, rather than file names).
Anyway, talk of "RPM dependency hell" is somewhat redundant now; with apt-get (yes, we know Debian had it first) and yum and, most importantly, robustly-buit packages (e.g. Red Hat's, and freshrpms.net), dependencies are resolved automatically.
--
From bitter experience, and from the last seven years working in computer security, I've learnt that for most things, there's at least one person in the world writing and releasing open software who understands the problem-space better than I. Rather than trying to reinvent the wheel, and learning all the snags and gotchas the hard way, I use their mature, battle-hardened code and get on with life. Of course, that doesn't prevent me checking return codes and using assert() regularly, as all paranoid programmers should.
Based on your comments, either you're in the top 0.1% of software authors out there, or you have an over-inflated sense of your own abilities. :-)
--
I have always had problems with Perl when I go to install a new module from CPAN if Perl was installed with an RPM file or came with the system (i.e. installed when the system was installed). Perl itself works great, but some CPAN packages barf when I try to compile them.
You want cpan2rpm
--
--
I make it your saving is under 300GBP, without taking account of the warranty, plane ticket/lost time from holiday or mailbox rental.
--
I don't know what it's like in the US, but here in the UK, the cost of new PCs is making PC "repairs" uneconomic if the repairer wants to charge rates similar to those of plumbers and the like (to put some numbers on that, a typical rate for a plumber is 60GBP per hour, and a new PC costs from 300GBP, with monitor and preloaded copy of whatever the latest flavour of Windows is; how much work do you reckon can do in under 5 hours?)
Of course, this does discount the stupid and the penny-wise-pound-foolish, whom are probably the best cash cows out there for any business.
--
Have you got any more information on that? That appears to be a serious vulnerability that I hadn't previously heard about... Securityfocus.com's vulndb doesn't seem to know about it, either.
--
I'm not a great fan of FW-1, but it's a shame that whoever plunked down the cash for the software didn't also pay for some training for you; that snippet would have been covered in the basic course.
--