Unreal 2004 has over 2 million lines of code. Monkey island (~1989) had in the viscinity of 50 000. Thats a big difference;
Sure is. Do you have a reference for that 50 000 lines of code estimate for Monkey Island? And do you know if that included both the engine code and the game/story code (which were probably written in different languages, as far as I've heard).
I'm not disputing your figures, you understand, I'd just like to know where you got them from.:)
Workflow Issues - What do large projects do to cut down on complexity and compilation time? They have interfaces. The hardest thing about game development is making the complex code run *fast*. That means no stupidly complex interfaces - every cycle counts. You often want your modules to talk directly to each other.
Wow. That sounds like it contradicts every bit of guru advice I've ever heard on optimisation, most of which boils down to "optimising too early is a mug's game" (pun intended). Or, to quote Knuth more directly: "Premature optimisation is the root of all evil." The "performance" section of these so-called pearls of wisdom:) is particularly appropriate.
And while searching for one of the above quotes I found this article, which explains the concept even better.
In any kind of programming, if you start optimising too early, you're almost certainly just going to be wasting your time (and massively fucking up the system design, too, if you're coupling modules together unnecessarily, as suggested by "You often want your modules to talk directly to each other.").
Well-designed interfaces also make it much easier to profile and do effective optimisation - when it's appropriate to do so, ie. when it's all working correctly, just too slowly.
Welcome to the game industry. Please leave your sanity at the door.
Sounds like it, yes. You have my sympathy.:)
Pete, who is very glad he isn't involved in the game industry, as he already has precious little sanity left from other industries...
I could easily imagine a productive environment based around GNU
screen and a terminal-based editor, mail client, news client, and IM
client. Throw in something like w3m, and other for images, its good.
Yup, that's pretty close to the way I've worked for most of the last
year or two. For me it's screen, of
course, along with:
P2P client - mutella, though I think
there are console options for other protocols,
IM/IRC client - irssi along with
the fantastic bitlbee (and if you
haven't heard of bitlbee before, take a look).
...and then I use good 'ol ratpoison for my window manager in
X for the occasions that I need graphics (ie. some web browsing,
viewing PDFs, playing graphical games).
Strike that. In most cases, multi-tasking can be very
counterproductive. Shell escapes and $EDITOR_OF_CHOICE is good enough.
It varies:-), though I agree generally speaking. I'm using KDE3.2beta
at the moment for a bit of a change, though most of the action is
still inside my screen(1) terminal(s). You do tend to (or at least
I tend to) find yourself more productive when you don't have
stray graphical bits and pieces around the place to distract you.
Of course if you need the GUI for your normal working
environment (ie. you're developing a GUI app), then, well there's not
much you can do but live with it.:)
> Okay, it's gone up by $1000 per month. Still fine by you?
First of all, you know you are being absurd, don't you?
I was trying to make a point. October_30th was saying something to the
effect of "I want anti-spam blocklists (or at least SPEWS
specifically) to just go away, I don't care if my ISP bills go up due
to spam as long as noone ever rejects my email."
I was trying to illustrate that he only "doesn't care" when the bills
go up if they only go up by an amount he considers
small. But of course his statement(s) on the topic would sound much
less dramatic and impressive if he qualified them with something like
"well, as long as the price increase wasn't more than $15/month or 5%
of the base monthly fee (whichever is less)."
I did think I'd explained this in the paragraph immediately following
my "$1000 per month" line, but apparently not.
How about prossecuting spammers and making them pay?
That'd only help (to some degree) for countries where there's an
effective anti-spam law and the spammer in question actually
resides in the country.
At the moment, the USA (where most of the major spammers live and run
their operations) doesn't have too many effective anti-spam laws in
place - and even the effective laws at state level were (AFAIU) badly
blunted by the federal pro-spam legislation recently passed.
Pity too, because the Washington anti-spam law (I think it was
Washington) apparently had some real teeth in it, and people were just
starting to realise how they could use it to hurt spammers.
Anyway, over a decent period of time, anti-spam legislation should
help to make a dent in the spammer industry in some countries
(mainly the USA). But spammers will still be able to operate
unhindered by the law in other countries. I'm just trying to say that
while some decent legislation (and enforcement of that legislation)
would be a good start, it won't be enough to solve
the spamming problem just by itself.
[...] but most ISPs who use a RBL do so without having much of a clue [...]
Some do, quite probably. But most?? Have you got any
kind of reference for that statement, or were you just extrapolating
from your immediate experience with a handful of RBL-using ISPs?
Sounds like a typical case of pot and kettle to me..
Maybe it is. Or then again, maybe it isn't. I guess those people still
bothering to read this thread (hi guys!) can make up their own mind.
[ I originally said: "No ISP that pays attention to abuse reports (and
acts on them) should ever end up in a SPEWS blacklist." ]
Your wording is exactly on the spot, SHOULD. Reality shows that it
does happen at times, and when it happens, that SPEWS is impossible to
reach to resolve the issue.
It's not impossible to reach them. It's just a one-way
communication channel (which many people seem to have trouble with,
and I can sort of understand that... I often wonder how many people
who bitch about SPEWS would still bitch if SPEWS did respond
to them). You post a message with the SPEWS record number in the
subject line to news.admin.net-abuse.email, with a clear and concise
explanation of the mistake, and it should (usually) get fixed pretty
quickly.
Otherwise it gets fixed when they notice the mistake, which also can
happen pretty quickly - and it probably can happen slowly as well,
though I haven't seen any examples of it taking longer than a week.
SPEWS make mistakes just like anyone else. They're human too.
Pete, you are completely and utterly missing the point that was being
made, and you actually have missed it repeatedly and consistently. I
suggest you go back and read and try to reply after thinking first,
and actually reply to the points being made in
The other thread you linked is quite right: to nanae, it's all a big game.
To some nanae denizens, it probably is. Others not. And still others, sometimes it is and sometimes it isn't.
Some of the nanae people actually do a fair bit of work fighting spam in its many and varied forms. I'm sure it can be quite tiring. So when they do their daily check of nanae and see another dozen posts from people saying "waah, spews blocked me and I'm not a spammer" or "hey, we just need to replace SMTP, that'll fix spam" or "yoRe all nErds, get a l1fe".... or the occasional amusing quasi-legal threats... well, resorting to humour is one way to deal with it.
Your idea of decent and helpful is wilfully screwing up a free service, wasting several days of volunteer time, and refusing to answer a straight question?
The people that erroneously listed the Samba mailserver (SPEWS, or perhaps Osirusoft, I'm not sure) are completely different from the people who responded to Martin in the aforementioned nanae thread. Now you're muddying the waters a fair bit yourself. Try not to post when you're pissed off, it tends to screw up your logic circuits.
I don't give a fuck about who Spews choose to list or not [...]
...but you'll cheerfully go on badmouthing SPEWS to anyone who'll listen, substituting passion for reason? Thought so. *wry grin*
No... I don't know why you thought I was assuming that. I didn't assume anything. Feel free to check it again if you like. I just thought if he was going to badmouth SPEWS that it'd be nice if he provided something we could look up and check for ourselves. And he did. Good for him.
His initial post gave no indication of his involvment with the Samba team.
Using SPEWS is like controlling immigration by blowing up every boat that might be carrying an illegal before it gets to the harbor.
First of all, mailservers that reject mail based on blocklists are not killing that mail. They're just not accepting it - and that rejection is communicated back to the originating mailserver in a fraction of a second.
Second, mailservers reject this mail based on a very simple criteria - where it comes from. Nothing else. Hell, that's the only thing you can reject on, really. You can't accept the mail, analyse it with a mailfilter and then reject it. SMTP just doesn't work that way, sorry.
I should never have to leave my ISP because some asshat admin at my upstream decides they don't like the way a friend's ISP does business.
You don't have to leave, of course. And it doesn't matter in the slightest what some admin at your upstream, asshat or otherwise, thinks.
The only people that matter are the ones that run the mailserver you use (and of course their bosses who have authority over them and (should) take responsibility for their actions). Your upstream is utterly irrelevant.
I think you might have got two distinct issues confused here.
BTW - I do like the way you make your prejudices so very apparent in the way you describe things.
The anti-spam, SPEWS-using admin is an "asshat", while the rogue ISP that is taking money from spammers and helping them to spray crap all over the net is just... "your friend's ISP and the way they do business." Hm.
As regards your rights to receive email untouched by blocklist - well, I'm sure there's plenty of ISPs that provide such a service.
There may even be some that will offer you alternatives within their own network - ie. you can use the mailserver with RBL protection or without - up to you.
Free webmail providers, by and large, blow complete donkey sack and you know it. This is a cop-out.
If it's just a matter of sending a few emails to get around a blacklisting, Hotmail or Yahoo would seem to be perfectly sufficient. They work, they send email, virtually all mailservers accept email from them. That's all you need.
Though fastmail are an outstanding free (and pay-for) webmail provider, even by my normally very critical assessment.
Have a look if you haven't heard of them before.
Regarding people who would smarthost your email, I don't know what your specific needs are and I don't know where you're located and I don't know how much you're willing to pay, so it's kind of a difficult question to answer directly. Contact the small ISPs in your area and ask around at your local Linux user group. Ask your friends who run their own mailservers.
Heh. I've actually met Martin personally, about a year ago while he was at Linuxconf-AU-Perth. He gave me a synopsis of his talk about distcc, which I finally got around to trying last week to build KDE from CVS. A very very cool app.
Then I had an amusing
exchange with him on this very same topic, regarding this very same SPEWS listing:-). Unfortunately I got around to checking his final response too late, so when I realised who it was I didn't have a chance to say hi:).
So I'll tell you what I was going to tell him, which is just that I thought the nanae-denizens talking to him were actually quite decent and helpful. The main factor that added confusion is that a couple of the posters wrongly assumed the issue was about Datapipe, rather than an actual mistake on osirusoft's or SPEW's part (genuine mistakes are, seriously, quite rare).
If Martin had been more familiar with nanae or with the way SPEWS works, he might have been able to get that point across more clearly.
But still, none of the replies were actually rude (by normal Usenet standards). None of them really flamed him. When he responded again and further muddied the waters from the real issue by making it clear that he didn't like the way SPEWS operates, a few people responded to the points that he made
in a tone that could be read as mildlypatronising. I suspect Martin didn't take that too well:), but, well, that tends to be the way things go when you pop into a group and start telling people they're wrong - without really researching the topic first.
And you've misrepresented what the nanae denizens actually said - they didn't say "we're not going to tell you which ISPs are good/bad", they tried to explain that it's just not that simple. There's too damn many ISPs, and (especially for the really big ones) their staff and policies change around too frequently.
Then, of course, SPEWS does actually provide a list, of sorts. You can query IP addresses belonging to an ISP and see if anything turns up. Spamhaus also provides a list of sorts, arranged a bit differently. Then, of course, you can always do a search through nana.sightings or nanae itself for the ISP's name. It's just part of the overall research that you might think about doing before investing significant money in an ISP contract.
There are some in the rabid fringe in nanae who get pissed off and be assholes about almost everything. But the majority of nanae regulars are generally intelligent, polite and helpful people... as long as you're polite to them. Much like Usenet as a whole, really.:-)
You're certainly under no obligation to, but I'd be interested if you could post a link to the thread in question via Google Groups or something. If you did indeed get treated inappropriately, I'd like to make a note of the people who did so, so I can pay less attention to them in future when reading nanae.;-)
The "downstream" people who didn't receive mail due to a SPEWS listing - well, the deal should be that if it was just tagged, they could still receive it (and filter it out, if they chose). If it was rejected at the target mailserver, then the person who sent the mail should know that their mail didn't get through and should take alternative measures.
What I'm trying to explain is, for the situation you describe, the suffering is not that high and temporary workarounds are generally available. I mean, what happens on the other occasions that mail doesn't get through, for whatever reason (eg. a few hours of power blackout, a cable accidentally cut, etc.)? You temporarily resort to alternative means of contact. *shrug*
Re: the 2-year-contract-with-termination-fee situation - well, any such contract that had been vetted by a lawyer (and even many that haven't been) should have an escape clause along the lines of "if they're not fulfilling their end of the bargain, you can pull out. And may even be able to hit them with penalties for breaking the contract."
I can only presume that any long-term contract with an ISP would include something about the network service actually having to fucking work.;-)
Re: the screaming kids, I can only grin and say "you have my sympathies" - assuming it wasn't just a made-up example.:)
None of the major RBLs block entire ISPs for one idiot sending out spam. If the ISP terminates the spammer promptly upon receiving the abuse report, they may not even have one IP listed.
Hence it's not a case of "one idiot sending out spam from his comcast email" and the rest of comcast gets listed.
If you're finding your email rejected from some mailservers due to a SPEWS or Spamhaus listing, that's usually a sign that your ISP has been actively supporting and encouraging their pet spammer(s) over a significant length of time. It's a sign that they care more about their spammer than they do about you or any other of their customers now caught in the blacklist.
There are several good reasons that RBLs like SPEWS end up listing wide ranges of IPs in an ISP's netspace (rather than just the identified "spammer" IPs).
SPEWS stands for "Spam Prevention Early Warning System" - and bad ISPs will often shift spammers around to new IPs if the current IP gets blacklisted. Listing a wider range is basically saying "this ISP is a spam supporter and cannot be trusted - spam could start coming from any of their IPs."
Even if the spammer stays on the one IP address, just listing the spammer makes the spam-supporting ISP very happy - they can take the spammer's money and that of their normal customers. Listing more of the ISP's netspace is an indirect way to let the customers know what they're dealing with, and give them a reason to (a) pressure the ISP to dump the spammer, or (b) leave the ISP.
Even if the original problem spammer stays on the one IP and the other customers don't care about being listed, the logic is that no ISP is going to take on just one spammer. In for a penny, in for a pound. If they're going to take spammer money, they might as well take as much of it as they can. Hence you blacklist as much of their netspace as you can, because new spammers will likely start popping up everywhere.
Keeping track of single "spammer" IPs is a hell of a lot of unnecessary work for the blacklist manager. If the bad ISP is moving their spammers around all the time, it's almost impossible to maintain - it just turns into an endless game of whack-a-mole. Blocking the whole shebang is much simpler and easier.
Yeah, I agree re: the <10 word summary. I hope they take up your suggestion.
Otherwise:
So if "the ISP" with a problem is the only residential high-speed ISP in the geographic area, what do you expect all the other residential users in that area to do? Move house? Go back to dial-up?
Well, let me put it this way. I will never pay money to a spam-supporting ISP (once I know what they are). If I found out my ISP was knowingly hosting a spammer and refusing to terminate that spammer - bam. I'm outta there.
To put it another way - if you would stay with a spam-supporting operation just because they were the only high-speed residential option (and I'm quite sure many speed addicts would;-), then what would motivate you to leave them?
What if their staff were frequently rude, insulting, ignorant and dishonest/deceitful? What if they kept incrementally raising their prices and/or changing their bandwidth limits after saying they wouldn't?
What if their service just became really spotty and unreliable every few days? What would it take for you to just say "Fuck it, I'm going back to a halfway-decent dialup provider."?
The reason I ask these particular questions is because an ISP that's desperate enough for cash to sign pink contracts is an ISP that's desperate enough to cut corners anywhere it possibly can - ie. anywhere that won't be absolutely guaranteed to cause a mass customer exodus. Resulting in situations much like those I describe in the previous paragraph.
Don't reward asshole businesses by continuing to pay for their "services". Just say no.;-)
Look, there are several levels of problems here caused by spam:
1) Network's infrastructural problems due to heavy traffic caused by
spam. To be brutally honest, that is not my problem.
Erm. Well, yes it is when your monthly ISP bill goes up.
If it means that my monthly ISP bill will go up, fine by me.
Okay, it's gone up by $1000 per month. Still fine by you?
The point I'm trying to make here is that you're casually presuming
that "hey! doesn't matter, it's just a few bucks here or there." But
even if it's only five bucks a month - hell, even if it's only one
buck a month, that's money you're paying directly out of
your pocket to your ISP so they can deal with problems caused
by spammers. Money that you shouldn't have to be paying -
hell, you didn't cause the problem, why on earth should you be paying
for it?
And your ISP is supporting spammers.
And you don't see a problem with that????
If the ISPs and governments cannot help, nothing will. Vigilantism
like SPEWS will only help to speed up the fall of e-mail system
because it breaks down the means of communications deliberately.
The second sentence is just twaddle, so I'll ignore that.
But the first sentence - well... the fundamental idea of the SPEWS
form of social pressure is to persuade "good" ISPs to shun "bad" ISPs,
thus providing some form of punishment for bad ISPs that allow their
network to abuse others parts of the Internet. I mean, SPEWS is
nothing without the ISPs that use it to block email. And
those ISPs use it for a reason.
So there are ISPs that are helping to draw the line and say
"Anything beyond here is just bad behaviour, and we'll shun you," if
that makes you feel any better. Though it probably doesn't.:)
2) Spam in someone else's mailbox. Couldn't care less.
Somehow I'm not surprised:). Thankfully some people have a greater
sense of social responsibility than you, and do care enough
to do something about it.
BTW, it's wonderful to see the phrase "couldn't care less" instead of the nonsensical "could care less" as used by far too many slashdotters.:-)
Filter it or get a monkey to push the delete button, I don't care.
What I care about is that my legit e-mail gets delievered and received
by people.
"Other people's problems don't matter, only my problems matter. If
they have to continue suffering tidal waves of spam so that I don't
have to deal with a few rejected emails, so be it. I'm far too
important to waste valuable hours setting up an account with a new
ISP, or valuable minutes organising an alternate smarthost, or
valuable seconds setting up a backup webmail account."
Sigh. I mean, seriously dude. Listen to yourself. I'm exaggerating
(just slightly) for effect, but that's essentially what you're saying!
3) Spam I get in my mailbox. Sure avoiding the pure raw spamfeed is
nice, but less draconian filters can take care of it.
Heh. The filter-only crowd always say that. And for a small-scale
email address that isn't on every spammer database on the planet, it
might even be true. For one person. At the moment. Let us know how
well it's going in a year's time.
I'd rather have pure unfiltered, unscreened feed [...]
Absolutely fine, knock yourself out. If you want a pure,
unfiltered, unscreened, untouched mail feed, more power to you. Go for
it. I'm sure you can find an ISP that will provide such a feed.
[...] from an ISP that doesn't care if it signs up spammers [...]
...But you sure as hell don't need to go to a spammer's ISP to get it.
I think you've failed to grasp how many people were suffering from the "disease" of the spammer on your network. Those people no longer have to worry about the spammer on your network. The fact that you (presumably not a spammer) get your mail rejected from their network (along with the spammer) is not their problem. It's your problem, and you should bloody well make it your ISP's problem.
If you were recieving all the email sent out by the abuser on your network, you'd probably get a better perspective on the scale of the "disease" - and realise that the "cure" in question is a perfectly reasonable one.
BTW: you still have the choice to "simply" delete/filter the spam you receive;-).
And if you think finding and using a decent webmail provider is arduous, then... well... I think the word "arduous" must mean something very different in your part of the world.
On the contrary, I have seen Spews list machines essentially by mistake.
I did say that mistakes get made occasionally. Very occasionally. The line gets more blurred with SPEWS, because a lot of people think they're listed by mistake when they really aren't. You sound like you know what you're talking about though, so I'll take your word for it. But yeah, there's not much you can say. Shit happens. A week before the fix isn't really that bad - though it probably would have been much quicker if a post clearly specifying the mistake had been made to news.admin.net-abuse.email as soon as the error became apparent.
It's a bugger, yeah, but mistakes do happen.
BTW:
But the downstream users who lost a week's worth of email might feel differently.
Um. That doesn't make sense. If a machine was SPEWS-listed accidentally, the users of that machine don't lose incoming email - just some of their outgoing email would get rejected, very explicitly. They could resend it as soon as the block was lifted.
Unless I've misunderstood what you were trying to say, which is certainly possible:).
Webmail isn't perfect, no, but it's a pretty reasonable solution for most "ordinary" end-users.
By the way, you can get around the Yahoo-dropping-free-POP3 problem by using YahooPOPs, a handy little app that acts as an SMTP/POP3 gateway on your desktop machine - so you can effectively use free YahooMail with free POP3 again. If you want to.;-)
And there are other webmail places that offer free POP3/IMAP access without the Yahoo bullshit, specifically FastMail.
Your other responses... well, no offense meant (seriously:), but they really do sound whiny.
Come on, how hard is it to live without a broadband link? It's really difficult for me to muster up much sympathy for the "broadband or death" line when I've been on a 33.6k modem line for the last two years (yes, that's right - not 56k, 33.6k).
In any case, smarthosting really isn't much of a problem. You could swing a dead cat at your local Linux User's Group meeting and just about every person you hit would probably be happy to smarthost your mail for nothing. Though the usual technique is to find a clean ISP, pay them to handle it, then deduct that amount from your payment to your SPEWS-listed provider (as they're the ones that caused the problem and they aren't solving it for you).
I feel some sympathy for people in places like Brazil, where there is only one government-sponsored ISP (as far as I understand) which is a filthy sewer-pit of spam and thus blacklisted to the proverbial hell and back. They really don't have anywhere else to go. You do, though you're reluctant to admit it.
Look, it's nice that you think you have free choice, but the innocent
people who are on that list do not have any choice in the matter. And
the people they're trying to stay in touch with might also have no
choice but to use the list, if it's company policy, or if their ISP
uses it.
There's always a choice.
An enormous range of free webmail providers.
An enormous range of people that will smarthost for you at little
or no cost.
And of course, a fucking enormous range of alternative ISPs.
That's for the innocent people on the list, of course;-).
THIS IS A PROBLEM. You can claim it doesn't exist till the cows come
home, but it will still be there.
It may well be a problem. The important thing to remember, always, is
that it's their problem, not our problem. You can
lead a horse to water but you can't make it drink - and you can give a
person a wide range of options for dealing with a SPEWS listing, but
it is not your responsibility to solve the problem for them.
This is a perfect example of why you should never just arbitrarily
block email because it comes from an IP on a list. Instead, programs
like SpamAssassin are useful because they use blocklists as a factor,
one among many, in determining whether to treat a message as "spam".
The problem with just using SpamAssassin is that it's very
CPU-intensive. And when the spam's already got onto your mailserver,
has already cost you in storage space and bandwidth.
SpamAssassin is good as a second (or third) line of defense,
but an RBL is much cheaper from the
CPU/bandwidth/storage perspective - hence one or more RBLs is
preferable as a first line of defense.
The cool thing about RBLs is the wide selection. Are you happy to
block confirmed open relays? No worries.
Do you want to block all of South Korea, as you never recieve legit
mail from there? No worries.
Do you want to block known and thoroughly reprehensible spam gangs
that have been booted off three or more ISPs?
No worries.
And of course there's a variety of other
blocklists, all with their own published criteria and standards.
No one says which ones you have to use. No one says you have to use
any of them.
But the major point is, if you're a target of a blocklist,
there's a reason for it (assuming the list admins didn't make a
mistake, which does happen very occasionally). And there are always
ways you can deal with the listing, ranging from ignoring it to
smarthosting email to changing your mailserver IP.
SPEWS are absolutely consistent with their listing criteria, and
always have been. If you're not a spammer and you've been included in
a netblock listed by SPEWS in Level 1, it is always after
your ISP has been repeatedly warned and they've done nothing
about the problem spammer.
A SPEWS listing always starts with individual IPs. Beyond that point,
it's the ISP's problem.
I thought it rather amusing that all the classic old DOS games that the guy mentioned are in fact available as source ports for Linux (and probably most other OSs).
But in case you didn't know, you can get a very damn cool version of Scorched Earth in 3d here:
http://www.scorched3d.co.uk/ (Windows, Mac OSX, Linux, source, etc.)
I played it with a bunch of friends at a LAN party recently, several of whom had played the DOS Scorched Earth before (I hadn't). Fan-fucking-tastic game. 'Twas a very satisfying moment when I was the first to discover that you could buy mini-nukes as weapons... *evil grin*
Guy who had just been hit with the nuke (along with everyone else who just heard the explosion): "WHAT THE FUCK WAS THAT???!?!??!"
Sometimes talking about things in those terms can be an effective way to illustrate things. And sometimes not;-).
On-topic - I used tcsh as my interactive shell from about '94 to '96. It pissed all over bash back then (purely from the interactive perspective). However, AFAICT, bash has improved quite a bit since then and has probably been superior to the latest tcsh for a while - in fact, is tcsh being actively developed anymore? It's probably an evolutionary dead end, to put it mildly:).
Heh. When I'm really working on an X11 machine, I use ratpoison. I tell you what, it really helps you to focus entirely on the work and not get distracted;-).
BTW, in keeping with the subject line of this thread - did anyone else notice that in the article, the writer talks about installing KDE3, but then supplies a screenshot referring to KDE2.2/KOffice1.1?
If I didn't already think it a pretty weak article, I'd find that kind of a lazy screenshot substitution mistake to be quite amusing.:-)
Pete.
Re:Let the conspiracy theories begin...
on
Make More Mistakes
·
· Score: 1
fmaxwell sayeth unto the people:
I'm trying to be fair. My contention is that a company which just
develops and sells open source software is doomed to fail.
Yes, I realise that now. Although... well, you're sort-of-kind-of
trying to redefine the point of the argument in your terms, to the
point that it doesn't really become much of an argument anymore.
You're doing a bit of the No True
Scotsman argument, tangled up with a few other things.
What I think you're saying (putting it in my terms instead,
which is always fun:-) is that a company whose only source
of income is from the "sale" (see below) of open-source software (by
the OSI
definition) will not make enough money to cover their costs of
development, let alone make a profit. Well... yeah, that's probably
correct. But you don't even need to use the term "open-source"
software there, you can be much more generic. How about:
"anyone who tries to survive
only by selling stuff that they
also make available for free, and that anyone else can (a) duplicate
and redistribute freely, (b) enhance (or not) freely, and (c) sell or
redistribute at any price (including zero).... will fail."
True? Yeah, probably. Only you're not casting
nasturtiums-by-association at open-source software
businesses anymore when you put it like that:-).
On a related line - remember of course that referring to the "sale" of
software is a very loaded and misleading term. Microsoft sure
as hell don't "sell" you software (or so they'd like you to think),
they license you to use their software - under fairly limited
terms - in exchange for a fee (or so they'd like you to think;-).
And of course that doesn't just apply to Microsoft, it applies to just
about any software company. I'm not sure what the legal status of
software is in the USA at this moment, but I've heard enough
stories about eBay disallowing resale of software products to make me
think that consumer's rights with regard to software products are not
very well established.
Back to quoting fmaxwell again:
I was not attempting to troll (you can probably tell that from my
karma bonus and comment moderations).
Yep, I realised that after reading a few more of your comments. Just
to make it clear, I don't think you're a troll
anymore (I wasn't completely sure before). And you do argue your
points lucidly and consistently and politely - for which you deserve
praise, even if you are wrong.;-)
I don't consider it open source to demand a fee from a user based on
the intended use. That said, I'm not going to add it as a condition at
this point.
I must admit it was sneaky of me to stick that in:), but I thought
that that would probably fall under the category of special conditions
you forgot to add to your list.:-)
I'm really trying to emphasise the point here of what you seem to be
doing - you're narrowing the target so much that there's not really
any point in using the term open-source software
unless you're trying to do the guilt-by-association trick.
I mean, think again about what you're saying - you want examples of
companies for which:
their only source of income is through "sales" of their
open-source software products (see above about the meaninglessness of
the term "sale" wrt software, not to mention that any non-trivial
company almost has to have more than one source of
income),
they provide free downloadable copies of said software products
(ie. unlike the Kompany,
Slashdot Mirror
Sure is. Do you have a reference for that 50 000 lines of code estimate for Monkey Island? And do you know if that included both the engine code and the game/story code (which were probably written in different languages, as far as I've heard).
I'm not disputing your figures, you understand, I'd just like to know where you got them from. :)
Wow. That sounds like it contradicts every bit of guru advice I've ever heard on optimisation, most of which boils down to "optimising too early is a mug's game" (pun intended). Or, to quote Knuth more directly: "Premature optimisation is the root of all evil." The "performance" section of these so-called pearls of wisdom :) is particularly appropriate.
And while searching for one of the above quotes I found this article, which explains the concept even better.
In any kind of programming, if you start optimising too early, you're almost certainly just going to be wasting your time (and massively fucking up the system design, too, if you're coupling modules together unnecessarily, as suggested by "You often want your modules to talk directly to each other.").
Well-designed interfaces also make it much easier to profile and do effective optimisation - when it's appropriate to do so, ie. when it's all working correctly, just too slowly.
Sounds like it, yes. You have my sympathy. :)
Pete, who is very glad he isn't involved in the game industry, as he already has precious little sanity left from other industries...I think you slipped and made your personal prejudices a little bit too apparant there.
Pete.He only does that to be nice and polite. He's not legally obliged to.
Pete.Yup, that's pretty close to the way I've worked for most of the last year or two. For me it's screen, of course, along with:
It varies :-), though I agree generally speaking. I'm using KDE3.2beta
at the moment for a bit of a change, though most of the action is
still inside my screen(1) terminal(s). You do tend to (or at least
I tend to) find yourself more productive when you don't have
stray graphical bits and pieces around the place to distract you.
Of course if you need the GUI for your normal working environment (ie. you're developing a GUI app), then, well there's not much you can do but live with it. :)
Pete.I was trying to make a point. October_30th was saying something to the effect of "I want anti-spam blocklists (or at least SPEWS specifically) to just go away, I don't care if my ISP bills go up due to spam as long as noone ever rejects my email."
I was trying to illustrate that he only "doesn't care" when the bills go up if they only go up by an amount he considers small. But of course his statement(s) on the topic would sound much less dramatic and impressive if he qualified them with something like "well, as long as the price increase wasn't more than $15/month or 5% of the base monthly fee (whichever is less)."
I did think I'd explained this in the paragraph immediately following my "$1000 per month" line, but apparently not.
That'd only help (to some degree) for countries where there's an effective anti-spam law and the spammer in question actually resides in the country.
At the moment, the USA (where most of the major spammers live and run their operations) doesn't have too many effective anti-spam laws in place - and even the effective laws at state level were (AFAIU) badly blunted by the federal pro-spam legislation recently passed.
Pity too, because the Washington anti-spam law (I think it was Washington) apparently had some real teeth in it, and people were just starting to realise how they could use it to hurt spammers.
Anyway, over a decent period of time, anti-spam legislation should help to make a dent in the spammer industry in some countries (mainly the USA). But spammers will still be able to operate unhindered by the law in other countries. I'm just trying to say that while some decent legislation (and enforcement of that legislation) would be a good start, it won't be enough to solve the spamming problem just by itself.
Some do, quite probably. But most?? Have you got any kind of reference for that statement, or were you just extrapolating from your immediate experience with a handful of RBL-using ISPs?
Maybe it is. Or then again, maybe it isn't. I guess those people still bothering to read this thread (hi guys!) can make up their own mind.
[ I originally said: "No ISP that pays attention to abuse reports (and acts on them) should ever end up in a SPEWS blacklist." ]
It's not impossible to reach them. It's just a one-way communication channel (which many people seem to have trouble with, and I can sort of understand that... I often wonder how many people who bitch about SPEWS would still bitch if SPEWS did respond to them). You post a message with the SPEWS record number in the subject line to news.admin.net-abuse.email, with a clear and concise explanation of the mistake, and it should (usually) get fixed pretty quickly.
Otherwise it gets fixed when they notice the mistake, which also can happen pretty quickly - and it probably can happen slowly as well, though I haven't seen any examples of it taking longer than a week. SPEWS make mistakes just like anyone else. They're human too.
Erm... all of them?
To some nanae denizens, it probably is. Others not. And still others, sometimes it is and sometimes it isn't.
Some of the nanae people actually do a fair bit of work fighting spam in its many and varied forms. I'm sure it can be quite tiring. So when they do their daily check of nanae and see another dozen posts from people saying "waah, spews blocked me and I'm not a spammer" or "hey, we just need to replace SMTP, that'll fix spam" or "yoRe all nErds, get a l1fe".... or the occasional amusing quasi-legal threats... well, resorting to humour is one way to deal with it.
The people that erroneously listed the Samba mailserver (SPEWS, or perhaps Osirusoft, I'm not sure) are completely different from the people who responded to Martin in the aforementioned nanae thread. Now you're muddying the waters a fair bit yourself. Try not to post when you're pissed off, it tends to screw up your logic circuits.
...but you'll cheerfully go on badmouthing SPEWS to anyone who'll listen, substituting passion for reason? Thought so. *wry grin*
Pete.No... I don't know why you thought I was assuming that. I didn't assume anything. Feel free to check it again if you like. I just thought if he was going to badmouth SPEWS that it'd be nice if he provided something we could look up and check for ourselves. And he did. Good for him. His initial post gave no indication of his involvment with the Samba team.
Pete.First of all, mailservers that reject mail based on blocklists are not killing that mail. They're just not accepting it - and that rejection is communicated back to the originating mailserver in a fraction of a second.
Second, mailservers reject this mail based on a very simple criteria - where it comes from. Nothing else. Hell, that's the only thing you can reject on, really. You can't accept the mail, analyse it with a mailfilter and then reject it. SMTP just doesn't work that way, sorry.
You don't have to leave, of course. And it doesn't matter in the slightest what some admin at your upstream, asshat or otherwise, thinks. The only people that matter are the ones that run the mailserver you use (and of course their bosses who have authority over them and (should) take responsibility for their actions). Your upstream is utterly irrelevant. I think you might have got two distinct issues confused here.
BTW - I do like the way you make your prejudices so very apparent in the way you describe things. The anti-spam, SPEWS-using admin is an "asshat", while the rogue ISP that is taking money from spammers and helping them to spray crap all over the net is just... "your friend's ISP and the way they do business." Hm.
As regards your rights to receive email untouched by blocklist - well, I'm sure there's plenty of ISPs that provide such a service. There may even be some that will offer you alternatives within their own network - ie. you can use the mailserver with RBL protection or without - up to you.
Pete.If it's just a matter of sending a few emails to get around a blacklisting, Hotmail or Yahoo would seem to be perfectly sufficient. They work, they send email, virtually all mailservers accept email from them. That's all you need.
Though fastmail are an outstanding free (and pay-for) webmail provider, even by my normally very critical assessment. Have a look if you haven't heard of them before.
Regarding people who would smarthost your email, I don't know what your specific needs are and I don't know where you're located and I don't know how much you're willing to pay, so it's kind of a difficult question to answer directly. Contact the small ISPs in your area and ask around at your local Linux user group. Ask your friends who run their own mailservers.
That's what I'd do. *shrug*
Pete.Heh. I've actually met Martin personally, about a year ago while he was at Linuxconf-AU-Perth. He gave me a synopsis of his talk about distcc, which I finally got around to trying last week to build KDE from CVS. A very very cool app.
Then I had an amusing exchange with him on this very same topic, regarding this very same SPEWS listing :-). Unfortunately I got around to checking his final response too late, so when I realised who it was I didn't have a chance to say hi :).
So I'll tell you what I was going to tell him, which is just that I thought the nanae-denizens talking to him were actually quite decent and helpful. The main factor that added confusion is that a couple of the posters wrongly assumed the issue was about Datapipe, rather than an actual mistake on osirusoft's or SPEW's part (genuine mistakes are, seriously, quite rare). If Martin had been more familiar with nanae or with the way SPEWS works, he might have been able to get that point across more clearly.
But still, none of the replies were actually rude (by normal Usenet standards). None of them really flamed him. When he responded again and further muddied the waters from the real issue by making it clear that he didn't like the way SPEWS operates, a few people responded to the points that he made in a tone that could be read as mildly patronising. I suspect Martin didn't take that too well :), but, well, that tends to be the way things go when you pop into a group and start telling people they're wrong - without really researching the topic first.
And you've misrepresented what the nanae denizens actually said - they didn't say "we're not going to tell you which ISPs are good/bad", they tried to explain that it's just not that simple. There's too damn many ISPs, and (especially for the really big ones) their staff and policies change around too frequently.
Then, of course, SPEWS does actually provide a list, of sorts. You can query IP addresses belonging to an ISP and see if anything turns up. Spamhaus also provides a list of sorts, arranged a bit differently. Then, of course, you can always do a search through nana.sightings or nanae itself for the ISP's name. It's just part of the overall research that you might think about doing before investing significant money in an ISP contract.
And this advice specifically is extremely worth your attention.
As to the other links you provided - well, all I can say is that it's Usenet, for $DEITY's sake. Turn your sense of humour back on. :-)
Pete.There are some in the rabid fringe in nanae who get pissed off and be assholes about almost everything. But the majority of nanae regulars are generally intelligent, polite and helpful people... as long as you're polite to them. Much like Usenet as a whole, really. :-)
You're certainly under no obligation to, but I'd be interested if you could post a link to the thread in question via Google Groups or something. If you did indeed get treated inappropriately, I'd like to make a note of the people who did so, so I can pay less attention to them in future when reading nanae. ;-)
The "downstream" people who didn't receive mail due to a SPEWS listing - well, the deal should be that if it was just tagged, they could still receive it (and filter it out, if they chose). If it was rejected at the target mailserver, then the person who sent the mail should know that their mail didn't get through and should take alternative measures.
What I'm trying to explain is, for the situation you describe, the suffering is not that high and temporary workarounds are generally available. I mean, what happens on the other occasions that mail doesn't get through, for whatever reason (eg. a few hours of power blackout, a cable accidentally cut, etc.)? You temporarily resort to alternative means of contact. *shrug*
Pete.Re: the 2-year-contract-with-termination-fee situation - well, any such contract that had been vetted by a lawyer (and even many that haven't been) should have an escape clause along the lines of "if they're not fulfilling their end of the bargain, you can pull out. And may even be able to hit them with penalties for breaking the contract."
I can only presume that any long-term contract with an ISP would include something about the network service actually having to fucking work. ;-)
Re: the screaming kids, I can only grin and say "you have my sympathies" - assuming it wasn't just a made-up example. :)
Pete.None of the major RBLs block entire ISPs for one idiot sending out spam. If the ISP terminates the spammer promptly upon receiving the abuse report, they may not even have one IP listed.
Hence it's not a case of "one idiot sending out spam from his comcast email" and the rest of comcast gets listed.
If you're finding your email rejected from some mailservers due to a SPEWS or Spamhaus listing, that's usually a sign that your ISP has been actively supporting and encouraging their pet spammer(s) over a significant length of time. It's a sign that they care more about their spammer than they do about you or any other of their customers now caught in the blacklist.
There are several good reasons that RBLs like SPEWS end up listing wide ranges of IPs in an ISP's netspace (rather than just the identified "spammer" IPs).
- SPEWS stands for "Spam Prevention Early Warning System" - and bad ISPs will often shift spammers around to new IPs if the current IP gets blacklisted. Listing a wider range is basically saying "this ISP is a spam supporter and cannot be trusted - spam could start coming from any of their IPs."
- Even if the spammer stays on the one IP address, just listing the spammer makes the spam-supporting ISP very happy - they can take the spammer's money and that of their normal customers. Listing more of the ISP's netspace is an indirect way to let the customers know what they're dealing with, and give them a reason to (a) pressure the ISP to dump the spammer, or (b) leave the ISP.
- Even if the original problem spammer stays on the one IP and the other customers don't care about being listed, the logic is that no ISP is going to take on just one spammer. In for a penny, in for a pound. If they're going to take spammer money, they might as well take as much of it as they can. Hence you blacklist as much of their netspace as you can, because new spammers will likely start popping up everywhere.
- Keeping track of single "spammer" IPs is a hell of a lot of unnecessary work for the blacklist manager. If the bad ISP is moving their spammers around all the time, it's almost impossible to maintain - it just turns into an endless game of whack-a-mole. Blocking the whole shebang is much simpler and easier.
Pete.Yeah, I agree re: the <10 word summary. I hope they take up your suggestion.
Otherwise:
Well, let me put it this way. I will never pay money to a spam-supporting ISP (once I know what they are). If I found out my ISP was knowingly hosting a spammer and refusing to terminate that spammer - bam. I'm outta there.
To put it another way - if you would stay with a spam-supporting operation just because they were the only high-speed residential option (and I'm quite sure many speed addicts would ;-), then what would motivate you to leave them?
What if their staff were frequently rude, insulting, ignorant and dishonest/deceitful? What if they kept incrementally raising their prices and/or changing their bandwidth limits after saying they wouldn't?
What if their service just became really spotty and unreliable every few days? What would it take for you to just say "Fuck it, I'm going back to a halfway-decent dialup provider."?
The reason I ask these particular questions is because an ISP that's desperate enough for cash to sign pink contracts is an ISP that's desperate enough to cut corners anywhere it possibly can - ie. anywhere that won't be absolutely guaranteed to cause a mass customer exodus. Resulting in situations much like those I describe in the previous paragraph.
Don't reward asshole businesses by continuing to pay for their "services". Just say no. ;-)
Pete.Okay... October_30th:
Erm. Well, yes it is when your monthly ISP bill goes up.
Okay, it's gone up by $1000 per month. Still fine by you?
The point I'm trying to make here is that you're casually presuming that "hey! doesn't matter, it's just a few bucks here or there." But even if it's only five bucks a month - hell, even if it's only one buck a month, that's money you're paying directly out of your pocket to your ISP so they can deal with problems caused by spammers. Money that you shouldn't have to be paying - hell, you didn't cause the problem, why on earth should you be paying for it?
And your ISP is supporting spammers. And you don't see a problem with that????
The second sentence is just twaddle, so I'll ignore that. But the first sentence - well... the fundamental idea of the SPEWS form of social pressure is to persuade "good" ISPs to shun "bad" ISPs, thus providing some form of punishment for bad ISPs that allow their network to abuse others parts of the Internet. I mean, SPEWS is nothing without the ISPs that use it to block email. And those ISPs use it for a reason.
So there are ISPs that are helping to draw the line and say "Anything beyond here is just bad behaviour, and we'll shun you," if that makes you feel any better. Though it probably doesn't. :)
Somehow I'm not surprised :). Thankfully some people have a greater
sense of social responsibility than you, and do care enough
to do something about it.
BTW, it's wonderful to see the phrase "couldn't care less" instead of the nonsensical "could care less" as used by far too many slashdotters. :-)
"Other people's problems don't matter, only my problems matter. If they have to continue suffering tidal waves of spam so that I don't have to deal with a few rejected emails, so be it. I'm far too important to waste valuable hours setting up an account with a new ISP, or valuable minutes organising an alternate smarthost, or valuable seconds setting up a backup webmail account."
Sigh. I mean, seriously dude. Listen to yourself. I'm exaggerating (just slightly) for effect, but that's essentially what you're saying!
Heh. The filter-only crowd always say that. And for a small-scale email address that isn't on every spammer database on the planet, it might even be true. For one person. At the moment. Let us know how well it's going in a year's time.
Absolutely fine, knock yourself out. If you want a pure, unfiltered, unscreened, untouched mail feed, more power to you. Go for it. I'm sure you can find an ISP that will provide such a feed.
I think you've failed to grasp how many people were suffering from the "disease" of the spammer on your network. Those people no longer have to worry about the spammer on your network. The fact that you (presumably not a spammer) get your mail rejected from their network (along with the spammer) is not their problem. It's your problem, and you should bloody well make it your ISP's problem.
If you were recieving all the email sent out by the abuser on your network, you'd probably get a better perspective on the scale of the "disease" - and realise that the "cure" in question is a perfectly reasonable one.
BTW: you still have the choice to "simply" delete/filter the spam you receive ;-).
And if you think finding and using a decent webmail provider is arduous, then... well... I think the word "arduous" must mean something very different in your part of the world.
Pete.I did say that mistakes get made occasionally. Very occasionally. The line gets more blurred with SPEWS, because a lot of people think they're listed by mistake when they really aren't. You sound like you know what you're talking about though, so I'll take your word for it. But yeah, there's not much you can say. Shit happens. A week before the fix isn't really that bad - though it probably would have been much quicker if a post clearly specifying the mistake had been made to news.admin.net-abuse.email as soon as the error became apparent.
It's a bugger, yeah, but mistakes do happen.
BTW:
Um. That doesn't make sense. If a machine was SPEWS-listed accidentally, the users of that machine don't lose incoming email - just some of their outgoing email would get rejected, very explicitly. They could resend it as soon as the block was lifted.
Unless I've misunderstood what you were trying to say, which is certainly possible :).
Pete.Webmail isn't perfect, no, but it's a pretty reasonable solution for most "ordinary" end-users. By the way, you can get around the Yahoo-dropping-free-POP3 problem by using YahooPOPs, a handy little app that acts as an SMTP/POP3 gateway on your desktop machine - so you can effectively use free YahooMail with free POP3 again. If you want to. ;-)
And there are other webmail places that offer free POP3/IMAP access without the Yahoo bullshit, specifically FastMail.
Your other responses... well, no offense meant (seriously :), but they really do sound whiny.
Come on, how hard is it to live without a broadband link? It's really difficult for me to muster up much sympathy for the "broadband or death" line when I've been on a 33.6k modem line for the last two years (yes, that's right - not 56k, 33.6k).
In any case, smarthosting really isn't much of a problem. You could swing a dead cat at your local Linux User's Group meeting and just about every person you hit would probably be happy to smarthost your mail for nothing. Though the usual technique is to find a clean ISP, pay them to handle it, then deduct that amount from your payment to your SPEWS-listed provider (as they're the ones that caused the problem and they aren't solving it for you).
I feel some sympathy for people in places like Brazil, where there is only one government-sponsored ISP (as far as I understand) which is a filthy sewer-pit of spam and thus blacklisted to the proverbial hell and back. They really don't have anywhere else to go. You do, though you're reluctant to admit it.
Pete.There's always a choice.
That's for the innocent people on the list, of course ;-).
It may well be a problem. The important thing to remember, always, is that it's their problem, not our problem. You can lead a horse to water but you can't make it drink - and you can give a person a wide range of options for dealing with a SPEWS listing, but it is not your responsibility to solve the problem for them.
Unless they're paying you to, of course. :-)
Pete.The problem with just using SpamAssassin is that it's very CPU-intensive. And when the spam's already got onto your mailserver, has already cost you in storage space and bandwidth.
SpamAssassin is good as a second (or third) line of defense, but an RBL is much cheaper from the CPU/bandwidth/storage perspective - hence one or more RBLs is preferable as a first line of defense.
The cool thing about RBLs is the wide selection. Are you happy to block confirmed open relays? No worries. Do you want to block all of South Korea, as you never recieve legit mail from there? No worries. Do you want to block known and thoroughly reprehensible spam gangs that have been booted off three or more ISPs? No worries.
And of course there's a variety of other blocklists, all with their own published criteria and standards. No one says which ones you have to use. No one says you have to use any of them.
But the major point is, if you're a target of a blocklist, there's a reason for it (assuming the list admins didn't make a mistake, which does happen very occasionally). And there are always ways you can deal with the listing, ranging from ignoring it to smarthosting email to changing your mailserver IP.
SPEWS are absolutely consistent with their listing criteria, and always have been. If you're not a spammer and you've been included in a netblock listed by SPEWS in Level 1, it is always after your ISP has been repeatedly warned and they've done nothing about the problem spammer.
A SPEWS listing always starts with individual IPs. Beyond that point, it's the ISP's problem.
Pete.I mentioned this above, but anyway - you might like to try out the open-source, cross-platform and very cool Scorched Earth 3d
Pete.
I thought it rather amusing that all the classic old DOS games that the guy mentioned are in fact available as source ports for Linux (and probably most other OSs).
But in case you didn't know, you can get a very damn cool version of Scorched Earth in 3d here: http://www.scorched3d.co.uk/ (Windows, Mac OSX, Linux, source, etc.)
I played it with a bunch of friends at a LAN party recently, several of whom had played the DOS Scorched Earth before (I hadn't). Fan-fucking-tastic game. 'Twas a very satisfying moment when I was the first to discover that you could buy mini-nukes as weapons... *evil grin*
Guy who had just been hit with the nuke (along with everyone else who just heard the explosion): "WHAT THE FUCK WAS THAT???!?!??!"
Pete. :-)
Sometimes talking about things in those terms can be an effective way to illustrate things. And sometimes not ;-).
On-topic - I used tcsh as my interactive shell from about '94 to '96. It pissed all over bash back then (purely from the interactive perspective). However, AFAICT, bash has improved quite a bit since then and has probably been superior to the latest tcsh for a while - in fact, is tcsh being actively developed anymore? It's probably an evolutionary dead end, to put it mildly :).
But anyway, bash is still not in zsh's league. ;-)
Pete.Heh. When I'm really working on an X11 machine, I use ratpoison. I tell you what, it really helps you to focus entirely on the work and not get distracted ;-).
BTW, in keeping with the subject line of this thread - did anyone else notice that in the article, the writer talks about installing KDE3, but then supplies a screenshot referring to KDE2.2/KOffice1.1? If I didn't already think it a pretty weak article, I'd find that kind of a lazy screenshot substitution mistake to be quite amusing. :-)
Pete.Yes, I realise that now. Although... well, you're sort-of-kind-of trying to redefine the point of the argument in your terms, to the point that it doesn't really become much of an argument anymore. You're doing a bit of the No True Scotsman argument, tangled up with a few other things.
What I think you're saying (putting it in my terms instead, which is always fun :-) is that a company whose only source
of income is from the "sale" (see below) of open-source software (by
the OSI
definition) will not make enough money to cover their costs of
development, let alone make a profit. Well... yeah, that's probably
correct. But you don't even need to use the term "open-source"
software there, you can be much more generic. How about:
True? Yeah, probably. Only you're not casting nasturtiums-by-association at open-source software businesses anymore when you put it like that :-).
On a related line - remember of course that referring to the "sale" of software is a very loaded and misleading term. Microsoft sure as hell don't "sell" you software (or so they'd like you to think), they license you to use their software - under fairly limited terms - in exchange for a fee (or so they'd like you to think ;-).
And of course that doesn't just apply to Microsoft, it applies to just about any software company. I'm not sure what the legal status of software is in the USA at this moment, but I've heard enough stories about eBay disallowing resale of software products to make me think that consumer's rights with regard to software products are not very well established.
Back to quoting fmaxwell again:
Yep, I realised that after reading a few more of your comments. Just to make it clear, I don't think you're a troll anymore (I wasn't completely sure before). And you do argue your points lucidly and consistently and politely - for which you deserve praise, even if you are wrong. ;-)
I must admit it was sneaky of me to stick that in :), but I thought
that that would probably fall under the category of special conditions
you forgot to add to your list. :-)
I'm really trying to emphasise the point here of what you seem to be doing - you're narrowing the target so much that there's not really any point in using the term open-source software unless you're trying to do the guilt-by-association trick.
I mean, think again about what you're saying - you want examples of companies for which: