Re: Register units catching on
> Where can I get a tape measure marked out in Olympic swimming pool lengths, widths and depths?
I can sell you one. For a smalle fee...
Vic.
5860 publicly visible posts • joined 7 Dec 2007
> I can't recall an instance where any manufacturer/retailer offered you the faulty bits back
It comes from things like car repairs - if you take your car into a garage to be fixed, and you *buy new parts to be fitted to it*, the old parts are still your property, and you have the right to have them returned[1].
This, of course, is an entirely different situation to what Apple are doing with warranty repairs.
Vic.
[1] You should generally exercise this right. I've seen garages swear blind to haev replaced parts that they clearly haven't touched[2]. Demanding the old parts back usually forces them to admit the lie...
[2] There was someone on the Citroen XM list a few years back that had taken his car into a repair place. They told him they'd changed the shocks. But XMs don't have shocks...
> Think about it, are they really going to send a £100 / hour lawyer to fight a £100 claim.
Royal Mail claim to have spent far more than that on the claim I made against them. They sent me a ream of paper (mostly legalese), including a breakdown of their costs. And my claim was for very much less than £100.
Once they'd figured out I'm not the sort to be intimidated by that approach, they sent me a cheque.
Vic.
> IBM never, ever, intended their PC to be an "open" platform; it took Compaq's clean-room reverse engineering
Well, they did supply an awful lot of information in the blue Tech Ref manuals. Like a complete source listing of the BIOS...
Vic.
> If I saw someone in a sysadmin team deploying a server configured using webmin
> then I would raise an appropriate alarm immediately
You need to be a little careful with that...
I frequently install webmin on new builds - not because I need it (although it's rather useful as a MySQL browser), but because I might not be there when something goes wrong. Talking someone through a webmin interrface over the phone is very much easier than talking them through a CLI...
Vic.
> I suggest postfix, which is much easier to set up
That's somewhat subjective; personally, I find sendmail much easier to set up than postfix, but that's almost certainly down to the fact I have far more familiarity with it than I do with postfix...
> BTW this would be useless with the webmin approach.
There is a webmin module available for postfix. I've no idea how well it works...
Vic.
> If you use tools that edit the .mc directly - or you enjoy going in and editing the .mc by hand
> - then do not use the sendmail module in webmin
That's pretty much what I keep telling you. And you keep telling me you know better.
> I was taught emphatically to never edit the .mc file in sendmail directly
I very much doubt you were told that.
You were almost certainly told not to edit the .cf file directly.
> I am repeatedly and forcefully told that I am never to do anything outside of M4.
So you *weren't* told not to edit the .mc file. Like I said, then...
> M4 is where configuration changes are "supposed" to be made, and so I make them there.
And if you use the sendmail module in Webmin, that statement is no longer true, as it alters the .cf file directly for a number of options. Hence the warning I keep maknig, and you keep telling me isn't important.
Vic.
> Sendmail module on webmin has "M4 configuration."
Yes. But if you edit that, it'll throw away stuff you've done with the other config tools, which edit the .cf file directly.
This is why I raise the issue. Every time, you tell me you get it - then post stuff like this.
Webmin is a fine tool, but it runs the risk of rolling back changes if you edit the .m4 file after you've used it to effect other changes. I wonder that you keep trying to ignore this very simple fact.
Vic.
> I doubt you'll find any serious Linux setup that isn't behind a dedicated firewall.
I can point you at a few thousand...
> I would advise turning SELinux off on your CentOS boxes.
I wouldn't.
SELinux is very, very effective. Russell Coker used to publish his root password on his website and let you shell into his machine to play with it. It was quite a stunning demonstration.
SELinux often needs to be disabled because the admin doesn't understand it well enough - and that's fine, it's still a fairly new technology. But it should be left enabled if at all possible, because it really does stop bad stuff happening.
Vic.
> Not sure I agree about using Webmin!
Webmin is a Good Thing(tm). It dramatically increases the discoverability of a server's features for the uninitiated.
But as I say every time the subject comes up, it has two significant problems: you need to be *very* careful if you try to have multiple users (as it doesn't really have them - they're all subsets of root), and you shouldn't do much sendmail administration with it (it writes the sendmail.cf file directly, meaning the sendmail.mc file gets out of sync, so future .mc modifications will roll back your webmin changes...)
Vic.
> Or those who use softfail in their SPF entries
SOFTFAIL is perfectly acceptable - it's easy to get something wrong the first time you try it.
You can use that in SA filters if you think that's a good idea.
But hard authorisation - that shouldn't be a reason to score a mail as spammy. +all is a disaster..
I'd like to see it removed from the standard, TBH. I can see no legitimate reason for it. Orthogonality be buggered.
Vic.
> I don't think I've *ever* had a spam from India
Do you want some of mine? They've been very prevalent lately.
Very few spams get through to me these days, but my filters[1] really aren't helped by the clueless fuckwits who insist on terminating their SPF records with "+all". For the uninitiated, that means "yes, absolutely everyone is explicitly authorised to send mail on behalf of my domain".
Grrr.
Vic.
[1] Yes, I know SPF isn't anti-spam. But my Spamassassin rules consider SPF authentication to be fairly indicative of hammy mail...
> they were stored, not going anywhere, therefore not 'in transmission'
Incorerct.
Subsection 2(7) of the Regulation of Investigatory Power Act 2000 [1] says :-
"For the purposes of this section the times while a communication is being transmitted by means of a telecommunication system shall be taken to include any time when the system by means of which the communication is being, or has been, transmitted is used for storing it in a manner that enables the intended recipient to collect it or otherwise to have access to it."
IOW, voicemail is explicitly *included* in the "in transmission" definition.
Vic.
[1] ::spit::
> I remember Deadrat in the early days
Your next phrase gives the lie to your memory stemming from "the early days"...
> no love from the developers because they were taking something essentially free and charging for it
RH were giving away a free version for years and years. CentOS only came into play when RH ballsed up the transition from RHL to Fedora.
Any anger at RH was because it *appeared* at the time that they were ditching the idea of a free distribution. They actually ended up doing something very different - hatching a free distro with community control - but we didn't know that at the time.
Vic.
> rpm -Uvh --force oracle-release-$version.x86-64.rpm
Not far off.
It does other things, though - like disabling your old repositories before it's downloaded the new -release RPM. Instant broken yum system if anything goes wrong.
And Oracle wonders why nobody trusts them...
Vic.
> What did I get wrong?
You confused Java the language with Java the VM.
Some parts of Android apps are typically programmed in the Java language, but Android handsets have no Java VM on them. They use the Dalvik VM, which is entirely different, being a register-based machine rather than the Java stack-based one. Think Z80 vs. 6502...
Vic.
> I have yet to see thin clients, virtualisation or any similar centralisation technology
> result in a reduced IT budget
It works quite well if you've got "outworkers" - users who control (and probably own) their own desktops, but expect flawless execution at all times, ro else they'll stop doing anything. But such people are generally only an issue for charities and other volunteer organisations.
For everyone else - I'm right with you. Lots of noise, lots of promises - but I've yet to see anything realised.
Vic.
> Paris has some amazing food, much of it well priced if you can find it.
I've yet to find any "well priced" food in Paris, but I've had some amazing nosh there.
I suspect the trick is to talk to the staff in French[1]...
Vic.
[1] No, there's no chance of my being mistaken for a native. But they do seem to prefer it...
> Why the willy waving about how far I've SCUBA'd?
It's not willy-waving. It's an attempt to get you to think about the depths you're talking about. Your "at most a couple of hundred feet deep" comment indicates that you have no understanding of the logistics of getting men to that depth and keeping them there for sufficient time to be able to do anything useful.
> I've done zero SCUBA diving.
I suspected as much. It might be a good idea to bear that in mind when discussing the viability of underwater attack at any great depth.
> I was pointing out that it would be possible to use SCUBA as proposed
And I was pointing out that it *wouldn't*.
> shall we keep trying to argue anyway?
Depends. Are you planning on continuing to talk complete tosh about diving?
Vic.
> If navy divers baulk at 90m dives, then they ain't navy divers!
Nobody mentioned baulking at anything.
What I'm trying to point out is that a 90m dive is non-trivial, and you don't just plonk divers in willy-nilly.
Trying to take out a mini-sub with using divers is a non-starter.
Vic.
> Don't mines detonate when something hits them?
That depends on the mine. Modern ones are somewhat more capable that the contact-spike things you see in war films.
Modern sea mines are typically looking for specific accoustic or magnetic signatures. This is why clearance diver kit is closed-circuit (to eliminate bubbling) and has all the chrome stripped off the brass bits (regs etc.)
CDBA is very, very expensive. But if you went in with a 5 grand unit like mine, it's pretty much guaranteed to blow up in your face.
There's bound to be a cheaper way to hunt mines and get them to explode than sending in $100K submarines. And it probably involves a Rasberry Pi...
Vic.
> the Persian Gulf and the straits are at most a couple of hundred feet deep
Max depth in the Straits of Hormuz is 90m.
90m is a non-trivial dive on Scuba.
> that's not going to be a problem
How many dives have you done to 90m? How many to " a couple of hundred feet", for that matter?
> Moreover, floating or moored mines will be at the surface, or within seventy feet of it.
But a submarine would not. It would approach at depth, then rise to meet its target. To do otherwise is to make it an easy target.
Vic.
> They are particularly good for FOSS, not so good for pay for shrink wrap.
That's incorrect. They are perfectly fine for shrink-wrap. Red Hat use exactly that model for their paid-for code.
> not so great if an update to a library breaks something else
And that's exactly why you use a repo by way of a package manager - dependencies are tracked.
> Most commercial linux users need to manually install pay for software
Your experience clashes with mine. And I manage commercial Linux systems for a living.
> Windows is able to host many different versions of software because it doesn't rely upon repos
Linux is also able to host many different versions of software, and it does (usually) rely on repos. That's because repos do not prevent multiple installations if that's what the user wants...
> You can also setup your own 'repo' for Windows
I suspect you don't yet know about the ways you can do this with Linux.
Vic.
> Or stego. in youtube posts.
Terrorspam. Stego in pictures of asian "doctors" trying to sell you little blue pills.
You could send it to the bloke that's supposed to be investigating the bad guys - if it gets through his spam filter, he'll undoubtedly bin it without looking at it.
Vic.
> we know this is not the correct figure but it is close because by its very nature it is a flawed calculation
That would seem to be a "best-effort" approximation.
> is that still fraud
Probably not.
The same section of the Fraud Act 2006 says :-
"A representation is false if—
(a)it is untrue or misleading, and
(b)the person making it knows that it is, or might be, untrue or misleading."
If the statemement, as above, is not misleading, then it would not fall under this Section.
Vic.
> I'd be 99% confident of identifying some banks who did the wrong thing
Isn't that a "Murder on the Orient Experss" moment? "They all did it".
> Is it worth £150 million to convict (say) 5 rogue traders
Yes.
The value of such convictions is not in getting some sort of revenge against those traders, it is to discourage others from following in their footsteps.
Vic.
> It is not fraudulent to say that someone else is doing a better deal than they are.
Yes it is.
Section 2 of the Fraud Act 2006 says :
"A person is in breach of this section if he—
(a)dishonestly makes a false representation, and
(b)intends, by making the representation—
(i)to make a gain for himself or another, or
(ii)to cause loss to another or to expose another to a risk of loss."
Vic.
> but it would do me absolutely no good, since I have neither the tools nor the expertise to take advantage of it.
This is incorrect.
Having source *does* do you some good, even if you're not a coder.
It gives you the possibility of taking that source to a coder to get things done to it.
So if a project is abandoned, you could arrange for your needs to be covered as you see fit. You have both the rights and the opportunity to do so. This cannot be said of a proprietary code, where you just won't often get the source, even if the author never wants to see it again.
There is obviously some cost in this route - but it's *your* decision whether or not the rewards warrant that sort of outlay.
Vic.
> In the case of statically linked binaries, the GPL is incompatible.
This is deliberate. The GPL sets out to copyleft all derivative works, and is generally successful This is a Good Thing(tm).
> The LGPL sort-of allows it -- you have to provide the binary blob as well.
This is not true. The LGPL permits linking against a proprietary blob. You only have to supply the LGPL work (as source).
> Often when companies do pass on the code, they do so under a NDA.
An NDA does not excuse the distributor from his obligations under (L)GPL. It also usually contravenes the "no additional clauses" rule (GPLv2 Section 6, for example).
> BSD looks close to what I want
BSD is a great licence, but it does not impose copyleft obligations. Given your misunderstanding of GPL, and your desire for copyleft, I'd recommend you look again at the GPL.
Vic.
> you can avoid a lot of the spoofing by using SPF
Sadly, there seems to be some sort of misinformation about SPF doing the rounds.
I've seen a lot of spam over the last few weeks sent from forged addresses. Looking up the SPF records for the domains in question, they all end in "+all". :-(
Vic.