I guess. What I don't understand is why there are no CVS clients that let you view all the files (ala VSS) as well as see if they are check out (and if so by whom). Followed by the in-ability to click to check out, and click to check in, CVS is just too frustrating. I made a custom VSS clone (imitates the gui, but not the file-structure) for my company due to the fact that all of us hated CVS.
I think you're confused about what the "concurrency" is about in CVS. Multiple developers are supposed to be able to simultaneously make changes to the same file.
Getting a list of the names of who has a file
"checked-out" wouldn't make a lot of sense.
Which is not to say that CVS doesn't have it's
problems (high on my list is the lack of support
for tracking a directory "mv")...
well i would blame the poster, because it sounds very different out of context versus what another person who replied to my post said.
actually, i tried not to be very harsh and didn't blame him, anyway, I just pointed out that it's a naive remark.
Look, that "beginning of the end" line was clearly a joke.
It was an intentional parody of the usual slashdot
buzz... and the really funny part about the joke is that you get it if you read the story, but you
don't get it if you dive into the discussion without knowing what you're talking about....
I don't know if anyone's noticed, but this is
essentially a pro-SCO article. If you hadn't read
anything else, you would think you needed to pay
SCO for every linux box in the house. It gives you
the impression that whether it's right or wrong,
SCO has a serious chance of winning in court.
Yeah, the author also has an anti-RMS bias, but
this "Booga booga, SCO is gonna *get* you linux
users" stuff is the absolute worst.
My problem with Perl is the ubiquitous use of the regular expressions.
Some people are good at learning stuff by heart. I am not
one of them. As a result, I've neither been good at
(bio)chemistry (that's why I became a physicist) or at
learning the syntax of regexps. To me Perl code looks like
line-noise you used to get when you used a modem on a bad
line. It's the same thing with Emacs' controls. There's just
no logic in it.
So, my question is: how the hell do you remember all those codes (perl, emacs, chemistry,...)?
This is actually a pretty interesting point. You're right
that it's a problem that computer engineering requires a
command of a vast number of largely (if not quite entirely)
arbitrary intellectual constructs, and I agree that the
sheer quantity of stuff you need to know is a difficulty.
In fact this is one of the reasons I haven't even bothered
to look at Ruby, PHP, or Python: I know Perl pretty well,
and despite claims that these other languages have some
stupendous advantages, they sound to me an awful lot like
just another way of doing things that I already know how to
do. (On the other-hand, learning more about the lisp-like
languages makes some sense to me, since they've genuinely
got a different approach to things...)
But to answer your question about how do you remember all of
that stuff, the only thing I can say is that you play with
it and get it to do things. To take your example of emacs,
yeah it doesn't make a lot of sense that you do a CNTRL p to
go the the "previous" line (why not u for "up"?), but if you
do it a lot the immediate feedback you get from the hitting
that keyboard gesture eventually burns the meaning of it
into your hindbrain.
Actually, I guess there's one more thing you can say:
there's often an underlying logic behind the design choices
that you start to percieve after awhile. Sometimes this
"logic" is just a matter of historical echoes, because the
current design will always imitate a previous design in some
respects. In the case of perl, it's regular expressions
have a lot of similarities to the regular expressions used
by other, earlier unix utilities (grep, awk, etc). It was
done that way on purpose: the idea was that if you were
familiar with unix, perl should be easy for you to pick up,
and that pretty much worked: perl is very popular with
unix admins. (Though as I understand it, it often goes the
other ways these days... people learn perl first, and then
discover later that they learned alot about unix.)
Thanks! In a testament to the power of caffeine and good friends, that chapter came together in the space of about four nights. I'd work regular business hours doing editing, then at 9 or 10pm I'd write the XML chapter. I got great feedback and clarifications from Matt Sergeant, Dan Brian, Michel Rodriguez, Adam Turoff, Robin Berjon, and other such Perl XML luminaries.
As you would have guessed if you heard me speak in 2000 [frii.com], I'm not the biggest XML user. I've mellowed since then, but I still don't do a lot of XML hacking. (One of the spare-time hacking things I've while here at O'Reilly, though, is to get our internal database of "what books are at what stage" into XML for easy grepping and reuse).
Heh. Well this may explain why you could write a
good, succinct guide to the subject. If you were a true believer you might have gotten lost in asserting that XML is the greatest thing since binary numbers (when really it looks like a slight improvement on CSV files)...
It was really funny how the XML hype kicked into
high gear long before the functionality was really
there... I remember checking in on what they were
up to in the mid-90s and feeling mildly shocked that they still hadn't settled on a way of dealing with namespaces.
Was it dropped because of a feud? It's great to be told why I do things:-) I don't remember it that way. The bit we deleted was just not a very useful observation. As you can tell, we tended to drop the parenthetical asides (when we remembered) [...]
Well okay, this is a case where I'm happy to hear
that I'm wrong. My apologies.
I can see why you'd prefer a simple program that
doesn't get in the way, but classic vi really is
too simple for serious work. It doesn't, for
example, do automatic saves as you go, so a
crash between saves is more likely to lose work.
I'm pretty sure that vim has features like that
these days, and certainly emacs has had them for
ages. (If you're interested in trying emacs,
the first thing you should do after running it
is ESC-x viper-mode to put it into a vi-emulator).
Yes, emacs has a lot of whizzy features that some
like to call "bloat", but it's actually not that
heavy-weight a program by todays standards, and most of the features are off by default. You can
turn off more of them by adding some lines to your.emacs file:
;Turn off GUI nonsense (scroll-bar-mode -1) (menu-bar-mode -1) (tool-bar-mode -1)
And this is probably recommended for writers:
;For displaying Latin-1 (ISO 8859) characters (standard-display-european 1);
What does it mean to say something like "RedHat's firewall is turned on by default"?
Once upon a time at least, a firewall was a separate machine that passed through selected network traffic, and was nearly crippled for any other purpose.
If you're talking about "turning on the firewall" on a given workstation, it seems to me that the meaning of the term "firewall" is drifting....
Yes, that's one way you hear the story. "oh,
Killdal was so irresponsible, he went off to
fly his plane, and left his *wife* to talk to
IBM!"
His wife was his lawyer. She always handled
negotiations with hardware vendors, e.g. that day she
had been dealing with HP before IBM showed up.
Killdal was the technical guy.
Maybe he knew IBM was coming and he dodged
them, and maybe IBM was looking for an excuse
to not use them, and the NDA bit was convienient.
Not bad, as far as I know, but you missed a
detail: the "small Seattle company" that
produced QDOS (Quick-and-Dirty OS) which
got turned into MSDOS (Microsoft Dirty OS?), had ripped off a lot of
their code from Gary Killdal's CP/M, produced
by Digital Research -- Killdal's name and
the name "Digital Research" were found in the
code... they lost the law suite, and Microsoft had to do a clean-room re-write
to avoid the tainted code...
One of the great mysteries in the story is why
*did* the IBM guys pick Microsoft -- which had
expertise only in languages like Basic, when
they could have licensed the leading microcomputer
OS of the time from Digital Research. There are
multiple, minor variations of the tale around,
one has it that the Digital Research guys balked
at signing the NDA, and wanted to negotiate other
terms. The Bill's Mom connection looks pretty
suspicious, in retrospect though. There's also
a rumor -- but only a rumor -- that Killdal had
been fooling around with someone's wife at IBM.
And you can continue the story further, e.g. Killdal
fights back releasing a new improved DOS called
Dr. DOS (pronounced Doctor Dos), which Microsoft
managed to supress with some classic dirty tricks
(spurious Windows errors to raise FUD about possible
incompatibility) until they could clone the features
in their DOS 5 release.
There's nothing wrong with pondering the biases
of a google-eye view of the world, the trouble with this article is that it just doesn't do
that good a job of it. It also might be better to focus on
practical techniques you can use to search differently, rather than act like you've discovered
some great flaw in google...
It seems to me that the real trouble with Google's
Pagerank is stunningly obvious: it's biased toward
the status quo of what everyone things is true,
or at least interesting. If you always start at
the top and work you're way down, by definition
you're always going to see the stuff that a large
body of people have approved of already. Is there
room in this system for new ideas to percolate up?
It might be interesting to start a "bottomfeeders"
weblog. Every week you pick a topic of interest,
google it, and start looking through the links that they've ranked down at the bottom.
There are other pretty simple techniques that might be worth a thought... yes it's annoying that
google won't let you search on punctuation (though this isn't just a google problem, in
the old days I always wanted to do an altavista search for site names with a tilde in them, to
find pages put up by individuals rather than
some slick dotcom...). But if you got a perl problem, you use google to find some perl sites,
and then use their search engines to look for info
on $| or whatever...
Oh, and has anyone given a thought to what's going
to happen when google finally goes public? How
long do you think the "Don't be evil" philosophy
is going to last, then?
Oh, come the hell on. A lot of this stuff I recognize from pre-IPO Netscape, which is to say
that you scraped up this list from a *really*
long period of time, and oddly enough despite all
this whacky-whacky stuff, many of these folks managed to
code up the original Netscape browser, transform
the planet, and retire millionaires...
Well, according to ex-mozilla employee list [ex-mozilla.org] one of the coders was:
- Driving an Alfa Romeo Spider, inspired by Dustin Hoffmans drive across the San Mateo bridge in "The Graduate", with a Netscape sticker
- Drinking 8 cans of soda a day and building a freakin replica of the golden gate bridge
- Doing bbqs at 5 Eden Avenue, Sunnyvale
- Kegs of guinness at above address
- Having the police turn up at above address - not to stop the party, but to check out Mike McQues Hummer
- 'Video conferencing' with parents back home in Ireland by sitting in front of Fish Cam!
- Heading with netscapees Tom Pixley and Rob Larrubio to Vegas to see U2 perform on the opening night on the Pop Mart tour, and getting more wasted than he has ever been in his life at 'Manhattan' in the New York New York hotel!
- Nerf gun wars.
- Duke Nuken wars.
- Mario Super Kart wars.
- Being interviewed or filmed once a week, and getting annoyed by it
- Writing a script that spat out random numbers on the screen for the film crews to get excited about
- Touring Be when they had 10 employees - and then getting a BeBox
- Taping up PABs monitor when he screwed up
- Beer Busts, and then going on the piss in Palo Alto with the cute admin girl from his building
Given that, I have to carefully prioritize where my money goes. Last year, I contributed to the ACLU, the EFF and to my public radio station, KQED. These are all good causes which, in my opinion, do demonstratively good things with my money and they all are tax deductible donations.
Well, if you're really asking us to prioritize
charities for you, I'd say you should dump KQED:
they're in fact pretty well funded, but have
a reputation for not doing anything worthwhile
for the money. At the very least you should
investigate the way they spend it... (kicking
in money to KPFA would make some sense, by
the way, presuming that you like things like
"Democracy Now").
Anyway, yeah, I have a similar list of priorities,
and while I'm not *tremendously* likely to give
to mozilla organization, I'll certainly keep them
in mind...
Albums are already a dead art form, they were
killed by CDs. It used to be the artist had
to think carefully about what was going to go
on the album, and where it was going to go:
a two sided LP is necessarily divided into
two "movements" of only about 5 songs each, so
there are two separate beginnings (and the first
track on side B was famously the "sweet spot" to
put an intended hit), and two conclusions.
All of this is lost in the CD versions of LPs
that most of you are used to listening to now:
the CD just plays straight through what was
originally the gap between side A and B, and
then after the original finale of the LP, there
are a bunch of typically very minor "bonus" tracks
tacked on that spoil the effect.
And as for music newly being released on CD,
the problem is no longer "which piece should I
choose to put on this album", but "how am I going
to fill up so much space?". It's rare for a
CD to be padded with actively bad tracks -- not
unless you're buying a really light weight artist --
but it *is* really common to have a CD that seems
a little samey, that gets a little boring before
the end. More often than not, I toss five CDs
in the carousel and play them on "shuffle":
sorry about your great work of art there, gang,
but I would've fallen asleep before the end
anyway.
...but corrupted by spectacularly awful 128k mp3 compression.
That being said, I am a subsccriber to eMusic and absolutely love it,
if only for the wide selection of non-mainstream stuff. I just wish
they had a clue in regards to mp3 compression.
Funny, the low bitrate has never bothered me, maybe because
I usually listen to them on cheesy PC speakers. But anyway,
they've got at least a clue... their on-line FAQ says:
"Due to high demand, EMusic is planning on increasing the
bitrate at which our songs are encoded."
I think someone else here was saying that the announcement
has just gone out...
They sent letters to the subscribers that were "downloading too much."
Not to people that were trading on P2P. In the end they sent letters
to about 70% of the subscriber base. They want your money but don't
want you to actually use the service.
I dunno what you're talking about. At one point Emusic had
some code running that looked through Napster offerings,
and emailed nasty notes to people who had put emusic MP3s
out on Napster.
Why you would need a MAC to get music quickly over the net either
means A) You are an idiot, B) You are an APPLE SHILL or C) You are
likely both
How about D) He actually likes the idea of some money going to the
artists, rather than stealing the music.
I can't comment on all the services mentioned, but Emusic
pays royalties on everything their customers download.
$10/month gets you full access to their collection of
MP3s, uncorrupted by any "DRM" (aka vendor hardware lock-in).
Can't wait for no DRM? That is like saying you can't wait until Best
Buy gets rid of those pesky cashiers. Why don't they just trust me to
leave an appropriate amount of money for the goods that I walk out of
the store with?? They are treating me like a criminal. Wah.
Emusic exists. You pay for a subscription, they give you
hassle free access to their entire collection of MP3s, and
they're straight, DRM-unencumbered MP3s. No problems with
backup copies, no problems with passing one to a friend.
If you try and re-distribute them on a mass basis you'll
probably hear from their lawyers eventually (as some Napster
folks found out a while back), but they don't harrass their
customer base just to make life harder for the small
percentage of ripoff artists.
(Score 5??? Someone boosted this AC garbage up to 5? WTF?)
At 99cents a track, this is a quick, easy impulse purchase for most
people, with instant gratification. Far easier than heading down to
the store and buying a CD... or ordering one and waiting for it to
ship.
A buck a track is incredibly expensive. You're paying
about as much as you would if you bought a CD, and the CD
would sound better.
Emusic *used* to sell music on a per track basis, but they
gave up on it because the subscription service was a lot
more popular with customers.
There's two kinds of businesses: one you try and give your
customers a good deal, in the other you learn how to
manipulate the press to pull in the total suckers
without a clue.
Just in case you're not up on the latest
Mozilla jargon, the scheme here is just to split it
up into a browser and a separate mail/news
client. Some already existing side-projects
are going to become the main development line.
(Took me a minute to figure this out...
Minotaur? Thunderbird? What?)
The point is that if you're just doing something simple,
like say dumping a database table to a file with code you've
written, which you're going to read later with code that's
also under your control, maybe you don't need the full-blown
XML business. But "Ostensbile Markup Language" is close
enough to XML that you can still parse it with some XML
tools, like the perl package XML::Simple.
Right. And just to make myself perfectly clear, that
was saracasm.
I didn't mean to imply that your summary of perl's
strength/weaknesses was a bad job. For one thing,
you point out that it is indeed possible to use
discipline in developing perl code, something that
many of it's detractors seem to miss.
And as for that "right tool for the job" philsophy, that's
another cliche I'm a little tired of, though I don't know
that I really want to get into it... briefly there
is no "right tool": e.g. your opinion that perl isn't the
best for numeric work probably has some technical merits,
but if you've got a shop full of perl programmers, and
it turns out you need to do some numeric crunching,
the "right thing to do" might be to use the "wrong" tool.
It takes some effort to write good code in Perl, because
it is so easy to write very bad unmaintainable Perl
code. You need to discipline yourself when writing Perl.
On the other hand, if you're writing in a *real* language
like Python or Java, bug-free, easily maintainable code
flows out of your finger tips like bullshit from the white
house.
Slashdot Mirror
Getting a list of the names of who has a file "checked-out" wouldn't make a lot of sense.
Which is not to say that CVS doesn't have it's problems (high on my list is the lack of support for tracking a directory "mv")...
Yeah, the author also has an anti-RMS bias, but this "Booga booga, SCO is gonna *get* you linux users" stuff is the absolute worst.
But to answer your question about how do you remember all of that stuff, the only thing I can say is that you play with it and get it to do things. To take your example of emacs, yeah it doesn't make a lot of sense that you do a CNTRL p to go the the "previous" line (why not u for "up"?), but if you do it a lot the immediate feedback you get from the hitting that keyboard gesture eventually burns the meaning of it into your hindbrain.
Actually, I guess there's one more thing you can say: there's often an underlying logic behind the design choices that you start to percieve after awhile. Sometimes this "logic" is just a matter of historical echoes, because the current design will always imitate a previous design in some respects. In the case of perl, it's regular expressions have a lot of similarities to the regular expressions used by other, earlier unix utilities (grep, awk, etc). It was done that way on purpose: the idea was that if you were familiar with unix, perl should be easy for you to pick up, and that pretty much worked: perl is very popular with unix admins. (Though as I understand it, it often goes the other ways these days... people learn perl first, and then discover later that they learned alot about unix.)
It was really funny how the XML hype kicked into high gear long before the functionality was really there... I remember checking in on what they were up to in the mid-90s and feeling mildly shocked that they still hadn't settled on a way of dealing with namespaces.
I'm pretty sure that vim has features like that these days, and certainly emacs has had them for ages. (If you're interested in trying emacs, the first thing you should do after running it is ESC-x viper-mode to put it into a vi-emulator).
Yes, emacs has a lot of whizzy features that some like to call "bloat", but it's actually not that heavy-weight a program by todays standards, and most of the features are off by default. You can turn off more of them by adding some lines to your .emacs file:
And this is probably recommended for writers:
What does it mean to say something like
"RedHat's firewall is turned on by default"?
Once upon a time at least, a firewall was a
separate machine that passed through selected
network traffic, and was nearly crippled for
any other purpose.
If you're talking about "turning on the firewall"
on a given workstation, it seems to me that the
meaning of the term "firewall" is drifting....
Yes, that's one way you hear the story. "oh, Killdal was so irresponsible, he went off to fly his plane, and left his *wife* to talk to IBM!" His wife was his lawyer. She always handled negotiations with hardware vendors, e.g. that day she had been dealing with HP before IBM showed up. Killdal was the technical guy. Maybe he knew IBM was coming and he dodged them, and maybe IBM was looking for an excuse to not use them, and the NDA bit was convienient.
One of the great mysteries in the story is why *did* the IBM guys pick Microsoft -- which had expertise only in languages like Basic, when they could have licensed the leading microcomputer OS of the time from Digital Research. There are multiple, minor variations of the tale around, one has it that the Digital Research guys balked at signing the NDA, and wanted to negotiate other terms. The Bill's Mom connection looks pretty suspicious, in retrospect though. There's also a rumor -- but only a rumor -- that Killdal had been fooling around with someone's wife at IBM.
And you can continue the story further, e.g. Killdal fights back releasing a new improved DOS called Dr. DOS (pronounced Doctor Dos), which Microsoft managed to supress with some classic dirty tricks (spurious Windows errors to raise FUD about possible incompatibility) until they could clone the features in their DOS 5 release.
There's nothing wrong with pondering the biases of a google-eye view of the world, the trouble with this article is that it just doesn't do that good a job of it. It also might be better to focus on practical techniques you can use to search differently, rather than act like you've discovered some great flaw in google... It seems to me that the real trouble with Google's Pagerank is stunningly obvious: it's biased toward the status quo of what everyone things is true, or at least interesting. If you always start at the top and work you're way down, by definition you're always going to see the stuff that a large body of people have approved of already. Is there room in this system for new ideas to percolate up? It might be interesting to start a "bottomfeeders" weblog. Every week you pick a topic of interest, google it, and start looking through the links that they've ranked down at the bottom. There are other pretty simple techniques that might be worth a thought... yes it's annoying that google won't let you search on punctuation (though this isn't just a google problem, in the old days I always wanted to do an altavista search for site names with a tilde in them, to find pages put up by individuals rather than some slick dotcom...). But if you got a perl problem, you use google to find some perl sites, and then use their search engines to look for info on $| or whatever... Oh, and has anyone given a thought to what's going to happen when google finally goes public? How long do you think the "Don't be evil" philosophy is going to last, then?
Anyway, yeah, I have a similar list of priorities, and while I'm not *tremendously* likely to give to mozilla organization, I'll certainly keep them in mind...
All of this is lost in the CD versions of LPs that most of you are used to listening to now: the CD just plays straight through what was originally the gap between side A and B, and then after the original finale of the LP, there are a bunch of typically very minor "bonus" tracks tacked on that spoil the effect.
And as for music newly being released on CD, the problem is no longer "which piece should I choose to put on this album", but "how am I going to fill up so much space?". It's rare for a CD to be padded with actively bad tracks -- not unless you're buying a really light weight artist -- but it *is* really common to have a CD that seems a little samey, that gets a little boring before the end. More often than not, I toss five CDs in the carousel and play them on "shuffle": sorry about your great work of art there, gang, but I would've fallen asleep before the end anyway.
I think someone else here was saying that the announcement has just gone out...
If you try and re-distribute them on a mass basis you'll probably hear from their lawyers eventually (as some Napster folks found out a while back), but they don't harrass their customer base just to make life harder for the small percentage of ripoff artists.
(Score 5??? Someone boosted this AC garbage up to 5? WTF?)
Emusic *used* to sell music on a per track basis, but they gave up on it because the subscription service was a lot more popular with customers.
There's two kinds of businesses: one you try and give your customers a good deal, in the other you learn how to manipulate the press to pull in the total suckers without a clue.
(Took me a minute to figure this out... Minotaur? Thunderbird? What?)
The point is that if you're just doing something simple, like say dumping a database table to a file with code you've written, which you're going to read later with code that's also under your control, maybe you don't need the full-blown XML business. But "Ostensbile Markup Language" is close enough to XML that you can still parse it with some XML tools, like the perl package XML::Simple.
I didn't mean to imply that your summary of perl's strength/weaknesses was a bad job. For one thing, you point out that it is indeed possible to use discipline in developing perl code, something that many of it's detractors seem to miss.
And as for that "right tool for the job" philsophy, that's another cliche I'm a little tired of, though I don't know that I really want to get into it... briefly there is no "right tool": e.g. your opinion that perl isn't the best for numeric work probably has some technical merits, but if you've got a shop full of perl programmers, and it turns out you need to do some numeric crunching, the "right thing to do" might be to use the "wrong" tool.
Interesting anecdotes. There's some bad news out there for cell phone addicts:
Hold the Phone? Radiation from cell phones hurts rats' brains
Certainly it's exceeded the expectations of the guy who wrote it and sold it to Gates for ten grand.