Of course, many people don't bother with foreign key constraints and they do such things programmatically. Which, to my mind, makes nothing easier. It simply makes the programmer responsible for data integrity, which is a recipe for trouble.
If there's a better way to do this in MySQL I'd love to hear it. I can't find it in the official documentation, but to be fair I stopped looking when I finally found an incantation that fucking worked. "INTEGER REFERENCES up_books(id)" certainly works, but it doesn't establish a constraint, which makes the whole exercise silly in my opinion. Why on Earth would anybody pick MySQL? I certainly wouldn't, but the server where this Web app currently sits is dropping its Oracle license and they've only installed MySQL.
(Oracle has its own set of oddities and quirks which annoy me, but at least I had confidence in the engine. To my mind, PostgreSQL is so far superior to both MySQL and Oracle for small-to-large database projects it's not even funny. Easy, fast and predictable. For certain enterprise setups I can see where Oracle stomps all over the free software options, and if Oracle were more reasonably priced I wouldn't be opposed to using it. But you have to be a sadist to want to do something non-trivial in MySQL.)
Mein Broder: So, in MySQL, when you exceed the maximum size of a TEXT column, does it throw an exception, or does it just truncate the data to fit?
Me: Well, it being MySQL, it will probably do something differently on Tuesdays than it does on the vernal equinox... but it probably will throw an exception and bitch about how you suck at data planning. Which is the proper thing to do, because who would want their database silently truncating data?
Mein Broder: In this case, I'd actually prefer it, 'cause otherwise I'd have to programmatically truncate it myself. These data aren't really that important, and truncating would be acceptable. It would be nice if I could be a lazy programmer.
Me: I think you're out of luck. But let's take a look:
MySQL Manual -- If you assign a value to a BLOB or TEXT column that exceeds the column type's maximum length, the value is truncated to fit.
Me: Astounding. Your desire to be a lazy, shiftless programmer has been facilitated by other lazy, shiftless programmers who have built the world's most rickety database management system.
Nice article on the perfect office, lots of ideas.
One thing that seemed to be consistent is the idea that individual offices are key. This is true--we do better with personal space than with too much openness. Sometimes, however, that's not possible. Cubicles are a lower-cost solution to real offices. Notice I said "lower cost", not "low cost". A fully tricked out cubicle runs about $8000, and that's not for the fancy stuff. The walls, furniture and installation is spendy. It's just spendier for dedicated offices, using traditional construction.
So, if you're in cubicle hell, how do you deal with the distractions? When I was working in a cube, I found the best solution was to have as a general policy that if I was wearing my Panic Hat, it meant that unless the building was on fire, it's better to not disturb me. I'd put on the Panic Hat, headphones, and work dilligently until the crisis was over. My co-workers were perfecly willing to comply. (I was the only nerd in a small department of a big company.) If you have a bunch of real assholes, this doesn't work, but assuming semi-normal co-workers, an official arrangement that deals with those times when you really, really need isolation can work nearly as well as an office door.
"This is Freya with OSDN marketing: she needs 1000 Slashdot readers to fill out some damned silly survey before they'll let her see sunlight again. The unspoken subtext here is that we really think that you're a bunch of lonely, tosspot wankers who'll do any damn fool thing if you think there's a chance for a gash to be thankful to you and sleep with you out of obligation. Even a kinda goofy-looking one who looks like she's about to lunge for your throat."
Perhaps OSDN should fire their marketing people, including Freya. If you haven't seen this Slashdot ad, don't compulsively click reload until you do. It's not worth it.
What's with Slashdot advertising? It's the most incestuous bunch of crap I've seen. If you can't sell an ad to a company like newegg.com where a click on a Slashdot ad will get you 10% off a hard drive, then you're complete shit at sales. Not that I've ever seen an ad ANYWHERE where a click will get you a discount, or a freebie of some kind. Take that back--I clicked on a Slashdot ad to get the free RSA algorithm t-shirt. Back, oh, what... 5 years ago?
While perusing the JOE v3.0 story, I began to notice a trend. When the comment indicated a fanatical, purblind ideology, the poster's UID was >5 digits. When the comment was reasoned, meaningful, and even if contrary to consensus had good reason for being that way, the UID was 4 digits, often less.
(This has the potential to be a masturbatory ego-fest, so I'll note that while I have a 4-digit UID, I don't post comments, and don't consider my input to be particularly scintillating. I'm talking about the other 4-digiters.)
Too much can be made of the length of your UID, but as I continue to read Slashdot I notice this more and more often. I also see fewer low-number-UIDs posting, period, while I see the comment totals increasing. This sounds like a writeup for No Duh! magazine, the journal of the obvious for the pedantic; but there it is.
We hear quite a bit about how Slashcode does this, or does that to prevent spammers, trolls and assholes from clogging up the works. There are a dozen dozens of Slash-alike codebases to do similar things. So far, I've seen none that actively work to elevate the discussion. It's not enough to just get a lot of comments. It's the quality of the comments that really make the difference.
(The best I've seen is probably photo.net, and their end-of-article comments. They generally tend to be pure opinion, often contrary to the article, but that's good.)
Of course, we'll never see a Perl script that uses regular expressions to strip out the shitty comments and leave the insightful ones. As far as I can see, all the community-moderated codebases simply enforce groupthink. To date, there is nothing quite as useful or as important as a real, live editor, who promotes good words and good thinking and deletes (yes, deletes) the crap. Perhaps it is time to reintroduce the classic newsman's editor to the Web--a singular vision that makes a site worthy to read rather than simply engaging in wankery of the most pointless kind.
In reference to this posting, where did all these fucking commies come from?
At stake is not the issue Free software (I don't use the "open source" moniker because it's stupid and an obvious marketing ploy [speaking of capitalism...]). At stake is the triumph of top-down totalitarianism in the form of suited beurocrats from the UN. The Newsforge article is rife with choice bits of rhetorical propoganda that would make Goebbels all red-faced and toeing the ground in shame.
Between January 2002 and December 2003, concern shifted away from the needs of those the WSIS was created to assist to those most responsible for creating the digital divide in the first place.
Bullshit. The "digital divide" is not caused by gremlins from Microsoft or hooded demons from the BSA. The "digital divide" is caused by regular old assholery by dictators and totalitarian governments who put a higher priority on collecting gold teeth from homeless widows. Microsoft only wishes it had that kind of power. They have yet to stem the tide of illegal copies of their software circulating in China, and they've got those narrow-eyed bastards in Beijing who could be bought off, if they really cared that much about software piracy.
The United States position, formed at the behest of the Business Software Alliance, CompTIA, and other organizations dedicated to maintaining the status quo and curtailing the growth of free software, is that no software development methodology -- closed and proprietary versus open source -- be recommended over any other.
Gosh, you mean they want to let software stand on it's own merit? What jackholes!
Look, chump, Free software can be great. It can also suck ass, and often does. Free software assholes, like yourself, remind me of a Judy Garland/Mickey Rooney movie. All the kids are glum because Mr. Dipschitz's bakery is going to close down, but Judy and Mickey save the day when they snap their fingers, turn to the camera and say, "I know! Let's put on a show! I mean, 'open source' it!"
To take your buggered argument and turn it around on you, do you mean that it's better for Juan and Kuan and Keshaun to learn how to edit a fucking dotfile rather than do a pointy-clicky thing to configure their webserver? When they fail to edit the sendmail.cf correctly, and the village mail server blows chunks; how is that a benefit to the hovel-dwellers?
To turn this policy position into a Free software rant takes quite a few logical leaps of Kryptonian proportion. Such as this:
Free software might help put an end to the Petri dish of poverty and ignorance which is the breeding ground for AIDS.
Ah, yes, the always useful
command. I use it often!
Spare me the drama, mama. Free software isn't nearly as helpful to dragging people out of repression as a HMMWV full of US Marines. It can be argued (and has been argued, actually) that ready availability of free-to-cheap handouts is as dangerous as free crack to gradeschoolers. There is a way for people to break the cycle of repressive poverty, and that way has never been, nor ever will be, the province of nerds writing code. Notice I qualified poverty with "repressive"--the Amish aren't rich by any stretch of the imagination, but they seem to be happy enough. There's not enough cash in Tahiti to buy a Superbowl ad, but they Tahitians aren't bitching about their lot in life. Quality of life is not determined by how much you know--it's determined by how free you are, and that includes the ability to choose IIS over Apache, if you want to.
And the comments were typical Slashdot fare, too. What, did kuro5hin's database fall over again? Oh, that's right, only lefties have the kind of free time to moan over Free software--everybody else is too busy making a living.
Okay, so he stopped using Windows in 1998. Fine. I know he thinks that makes him a "rebel" and an "iconoclast", or whatever rock-star quality he thinks lives in a OS CD--but really, it just makes him a tool. After all, Macintosh users have been Windows-free since 1984. Welcome to the party, Mr. Limo! And then he writes an entire article about "coming back" to Windows.
Not only did he stop using Windows in 1998, he froze his sense of humor right about that time as well.
Tedious jabs at Micro$oft or Bill Gate$, or how clunky Windows is--well, it's played. Played like a Barry White album at a 40-year-old bachelor's pad. Leaving aside the factual and/or stupid errors in the article, the rest reads like a Freshman level creative writing exercise. Ooh, Mr. Limo, your stunning use of satire, she is so witty! What's that you say? Windows isn't like SuSE? LoLz!!!!!1
Eat a dick. Using Linux doesn't make you better, it doesn't make you smarter, and it certainly doesn't help you to pick up the chicks. Hey, guess what--I haven't used Linux since 1999! I've been pure BSD and Macintosh, ever since my RedHat box was rooted through the (default installed) wuftpd. oH mY gOd LiNuKs Is TeH sUcK!!!!!!111 Can I get a "hell yeah" from the boyz down wit the daemon and those fine, fine BSD bitches!
Nothing is quite so annoying as the back-patting that mid-level Linux nerds do to themselves. Higher geeks can specifically point to reasons why a Free unix serves their needs. It's choirboys and fucktards who hang out at the Linux section of CompUSA, hoping passersby will notice how l33t they are that think an operating system will change the fact that they're dateless and lonely at 30.
Bah. Back to your cave, troll. If this is what OSDN pays Roblimo for, I'd like to apply for the position of Official Curmudgeon. I can probably get a thousand responses an hour from pale, friendless virgins who huddle around their Gentoo box like it's a pocket pussy, just by airing the truth: if you judge yourself or anyone else by the OS they use, then you're a slaptarded spockhole who should be phased out with a woodchipper. Calling Mr. Limo!
Lots of advice for "C&C", I thought. Of course, in a perfect world, C&C is the best option, so advising others to do so would be smart.
So why don't we have C&C Free software? Oh, a few things here and there are C&C, but most of the Free software world (as I see it) is made up of Q&D Perl atrocities, or worse.
There must be 100 different "web-forum" software packages, all of them Q&D (and solving the wrong problems, too). The few that are interesting (Scoop, Slashdot) are best used in very specific applications, but largely are incomplete (or bad) copies of USENET. Why hasn't somebody made a C&C web-forum software? Because, nobody wants to work on C&C projects. They're slow to start, slow to finish, and the payoff is just as questionable as a Q&D project, but it takes longer.
C&C is not sexy. Q&D is. Which is why the Internet is slowing grinding to a halt, as nobody seems to be able to do anything terribly useful with it because nobody wants to do it right.
Know what? Harry Potter and the Order of the Phoenix was pretty good. It did not disappoint, anyway: it was just like the other Harry Potter books.
But, sometimes I get the feeling that J.K. Rowling has Harry do things totally randomly, just so the plot can carry on. I'm not bent out of shape on this issue, but a few times during the book, I'm wondering, "Jeez, this isn't the first time something of this magnitude has happened, and Harry went to Dumbledore then--why doesn't he go now?" I'll buy rebellion, a bit, but not for the big things. Harry's all concerned at the beginning because Voldemort's running lose, and when he's experiencing dreams where he's sharing thoughts with Voldemort, he decides to be rebellious?
Okay, 10 for 10 on Griffyndor-level bravery, but 10 for 10 on Brainless Gittery as well. And, I didn't buy it.
And I think I've got the formula down: Harry goes to school. Harry finds/does/is something bad. Harry runs to Ron and Hermione to talk it over. They give stupid and/or good advice. Harry ignores the advice, something bad happens, something good happens, Harry talks it all over with Dumbledore, school ends.
Finally, I was totally unimpressed with the ending. After pages of buildup, we get an exposition that any child older than 9 figured out by the end of Book One? Harry Potter vs. Voldemort. Big Shocker, Film at Eleven! Okay, the niggly details were appreciated, but hardly worth the effort. The reason for sending Harry back to Privet was kind of lame, and Dumbledore's Big Failure was about as shocking as petting the cat. I was hoping for something of Darth Vader Is Your Father-level, but it was not to be.
However, fifty points to Gryffindor for the extra Fred and George Weasley coverage! And I dig the fact that school-age kids around the world are getting a E-Z-View Primer on How Government Is Not Your Friend, and You Can't Depend On The Media To Tell You Everything. Woo!
First, as many pointed out, the graphs are pathetically wrong.
Second, it's comparing some funny stuff. Okay, there's no real competition to AfterEffects, which this review relies heavily on, but Premiere is having it's lunch eaten by iMovie and FinalCut Pro. I'd like to see a comparison between the Apple products and Premiere--how fast to go from raw footage to final edit? How fast are basic rendered dissolves between shots? That would be interesting.
These graphs only prove that PCs make good AfterEffects boxes, and only at the rendering stage. It's good to know that you can save 4 minutes when you're rendering, but if you can put the project together 3 hours faster on a Mac, then the rendering time becomes a rounding error.
So, whatever. Adobe's being a dick. This is a new corporate strategy (maybe they've been talking to Quark?), but I don't think it's a good long-term one.
Great jumping horny toads, it's not like a simple journal isn't rat-simple to program. I use nothing more complicated than Emacs (which, I grant, is pretty complicated), but I just use it to write an HTML file.
Lordy, these kids with their fancy-ass do-funnies and object-oriented crap-throwers.
During this time, I'll be cutting/pasting the link to all my posts. This is a flat-out underhanded trick. I'm aware of the hatred that will be focused at me for doing so. Hopefully, I'll ameliorate the evilness when I promise to restrict my posting in current discussions to truly Insightful and Informative (or at least Funny) posts, and not one-liners. I'll try to make the added bit slide in with the sig as best I can. There might still be a BR and "--" in between, however.
I would like to get the attention of those with sigs turned off, and the casual non-logged in user as well. For you anti-blackout people, a representative ratio of those will also think I'm a kook and will join your side. So you shouldn't bite my head off--a nip or two oughta do ya.
I'm hoping here to dispose of some misconceptions and answer a few
questions I've seen asked and been asked. I will probably cause a whole
new set of questions and a new set of misconceptions, here. Such is
life. I should preface this by saying that I speak only for
myself--others have different reasons and different desires, and they
can speak with their own voice.
Why do you hate Slashdot?
I don't hate Slashdot. I've said it before, I love the community. If
I thought this would destroy Slashdot, I wouldn't have started it.
Then why boycott Slashdot?
First, it's not a boycott. A boycott implies a continued non-use of
services until a goal or demand is met. This is, as I called it, a
"blackout" of a temporary (and short) duration.
I felt that the best way to demonstrate the importance of comment
posters was to become part of the "other 97%".
82% read 10 or fewer pages a day. 15% read 30 or less. 3% read more than 30 a day.
That 97%, while providing page views and eyeballs, mostly do not
contribute to the site. The other 3% are the ones that do "cost" more in
DB and bandwidth resources, but they are also the ones that provide 99%
of the content here. That content gets spidered by Google (that leads to
search engine hits), builds a stolid community (that leads to press
attention), and feeds the churning machine of story submitters and
review writers building and bringing more comments and visitors daily.
But I (income) - C (costs) = P
If you believe you can reduce a socio-economic dynamic such as
Slashdot into a simple profit and loss statement, there are hundreds of
surviving dotcoms who would pay big money to see that equation.
If it were that simple, Slashdot could remove comments altogether and
be more profitable. The problem with that scenario is that Slashdot
would fade away soon after.
How do you know that?
I don't. It's my contention that is the case. That's part of what I
wish to show with the blackout.
So how will you know if you succeed? What is your standard of "I
That's a hard question. I will never know what the internals of
Slashdot are doing or thinking. They could be printing out my amateurish
screeds and using them as toilet paper.
If it is never said again that "half the visitors don't care" about
comments, that would be a win. The fact that half of the visitors to
Slashdot don't click-through to read comments is irrelevant--without
those comments, Slashdot wouldn't have that other 50% visiting at all.
Mostly, I will see if the attitude I perceived in the past continues into the future. If so, I'll know that my small protest was lame and ineffectual. If it seems to have generated some positive attitude changes, I'll consider it a win.
I guess you think Malda hates comment posters?
No. Malda and the other "editors" read and reply to comments quite
often, considering their busy schedules. They obviously care about them:
do they fully appreciate their value, however? The impression I get is
that the comment sections are a gift bestowed upon the laity from the
High Priests of Slash. My contention is that the reverse is more true.
Frankly we doubt that 3% will really pay us at all. Notice the venom posted
in this discussion: this comes largely from that very 3%. Its ironic that
those who profess to hate us the most also load the most pages;)
You hate Malda, huh?
No. I don't know him. The handful of emails I've traded with him over
the years have been cordial.
I don't like their appropriation of the title of "editor" for
themselves, I admit freely. That is why I usually scare-quote "editor"
when speaking of the Slashdot team. If they are editors, I'm a duck.
So you want Malda to be filtered through a PR zombie?
No. I quite enjoy being talked to like a regular person and not a
consumer. It is refreshing.
The flip side to that coin is, if you're going to be honest in your
feelings, don't be surprised to find your feelings challenged. Copping a
wounded attitude because everybody doesn't agree with you is kind of
Do you just want free subscriptions to Slashdot?
No thanks, I have a subscription already.
Then are you looking for an apology?
The mental image I have of Malda doesn't allow for him to
apologize. At least, not a real one--"I'm sorry you're such a dork. Get
over yourself! There, I apologized!"
In addition, I'm not an apology-type guy. Some people may have been
happy that Bill Clinton apologized for slavery, but I think it's
stupid. Words are important, but apologies are words without
Really, this isn't about assigning blame--it's about fixing a
problem. I believe that the Slashdot crew have a misplaced notion of the
importance and true cost of comments. They don't hate or dislike them,
they simply misunderstand their purpose.
I had a similar experience on E2--I did a writeup for a node that I
happened across. I did not go through the "Writing a perfect node"
procedure--I saw a hole and filled it. For my efforts, one of the E2
"editors" lambasted me for not jumping through their hoops. I responded
that to jump on the heads of casual noders is probably not a good
plan. They (casual noders) are not the bulk of the E2 community, but
they are an important part of the "ecosystem" there. He responded with a
"love it or leave it" statement.
There is a choice to be made--is the core community important or not?
If not, then spare us the trolls, flooders and PWPs and dump the comments. If so, then walk and talk like
you know it.
Boy, your ego is sensitive.
Perhaps. We can talk about my ego some other time, though. That is
(at best) a side issue.
If you don't like it, why don't you just leave? Or start your own
The same reason I don't leave the US to start another country if I
don't like what's going on here--it's easier to try to change the system
than to start anew (most of the time).
And who's to say that another community won't gain my
audience, rather than Slashdot? Perhaps, if Slashdot continues on its
path and doesn't improve relations to its community, I will go. And
perhaps others will as well. Not in a bunch, mind you, but in dribs and
drabs, until only the trolls and crapflooders are left. As more people
feel marginalized, they will contribute less and less, feeling that
since "half the visitors don't care", they may as well not give their
opinions or ideas a voice.
At that point, Slash will have excellent triggers for detecting
crapflooders and bots, scripts and lamers; and somebody will take that
and use it to build a more stable (and more attuned) community.
I'd rather that not happen, myself.
So you maintain that comments and comment posters have value? Why
don't you go sell it, then?
Corn has great value. But not to another corn farmer. The comments
are a product of the value-producing community. I suppose you could
equate the community with the land, and comments as the crop. It's a bit
of a weak analogy, but let's run with it.
The land is filled with potential value--but in and of itself, it is
worth little. The comment community, which makes up the heart of
Slashdot, isn't worth much if they aren't producing. What they produce,
you can't shop around to other web sites--they have their own land and
corn stalks already, thank you.
(I thought about extending the analogy to the community equating to
corn farmers, but then I figured somebody would say, "But corn farmers
wear overalls and drive tractors! Your analogy stinks!", or something
equally pointless. So I'll just leave my pointless analogy where it
I still don't get it. Why are you doing this, and what do you
I've done my best to explain. Read what I've written again and then once more. If you still don't get my position, just ignore me. If you do get
it and disagree, don't participate. If you get it and agree, remember to
stop posting on Apr. 21-27th. Don't post comments, don't submit stories;
just visit the front page for the links. Become as the other
And, if you send me $1000 via PayPal today, I'll send you a real
prayer cloth that I used to ritually clean the sacred scrotum. Put your
hands on the screen, and let me heal you! Hallelujah!
Saved for posterity, from a discussion about the new subscription system:
... while I don't mean to dismiss the value
of comment posters, the percentage of readers
that read comments is small. Yes comments draw
readers, and keep them coming back. But half
of readers don't care! An accepted story
submission provides a benefit to hundreds
of thousands of Slashdot readers. A Score:3
comment is read by 1/50th of that. So if we
decide that an accepted story submission is
worth 1000 page views, you would need to post
perhaps 50 Score:3 comments to affect the
same number of people:)
For the record, my feelings on the Slashdot Subscription Embroglio rests firmly in the uninterested. I have almost zero opinion on the final outcome of subscriptions. I love Slashdot, and will probably subscribe at some point to support the site, but the details are dull (to me).
Says Rob Malda, "... while I don't mean to dismiss the value
of comment posters, the percentage of readers
that read comments is small. Yes comments draw
readers, and keep them coming back. But half
of readers don't care!" In that case, Slashdot would be much better served by dumping the flaky and irritating overhead of a DB server and filling the pipe with a longer "Favorites" list--which, essentially, is what Slashdot is once you strip away the comments and comment posters. This is where a meaningless SQL query puts dangerous statistics in the hand of the ignorant. If Malda thinks that he can divine real knowledge from a SELECT query, he is sadly mistaken. While I do not doubt the validity of the numbers, I seriously doubt the validity of his extrapolation of the data. The ebb and flow of a community cannot be read from the tea leaves of an Apache log file.
This easy dismissal of the value of the only providers of interesting and insightful content on Slashdot is offensive. Thus, I propose a small revolt. The (Hopefully) Great Slashdot Blackout.
T(H)GSB will be during the week of April 21 through April 27. Easy to remember, the full moon in April falls on the 27th. During that time, I will not be posting, nor will I click through to read the comments from the home page. I will become as Malda's idea of the typical Slashdot reader. I will provide no new content (neither comments, nor story submissions--although I'm not much of a story submitter).
During that week, I'd like to see if Malda sees Slashdot become a better place, or if it becomes the Hallowed Shrine of Troll. I'd like for the logs to be revisited and new queries run. And, I'd like for the "editors" to really see what the true value of Slashdot is--not the sum of click-throughs and page-views, but the sharing of knowledge and dissemination of information; the passing of experience from the more to the less.
This is where the (Hopefully) comes in. This is only meaningful if enough free content-providers (i.e., comment posters) agree to go along and participate. If there is only me and a handful of others who cease normal activities during that week, it will be pretty meaningless. Barely a dent will be made, and Malda and the other "editors" will never realize the incredible value they receive from comment posters.
To spread the word, I'm changing my sig to link to this journal entry. If you would like to help, you can link to this journal from your own sig, or you can simply resolve to enter into a voluntary one-week blackout. Pass the word. This will only work if a goodly number of comment posters participate.
To summarize, if you wish to participate, during the week of April 21 through April 27
Do not click through from the home page to the comment page
Do not post any comments to stories
Do not submit new stories
A useful HTML link to this journal entry (69 characters, should fit in most sigs). You'll probably have to unfungle it after the lameness filter gets through with it: