I should probably clarify that "taking matters into your own hands" doesn't mean you have to ditch Rails. You'll just write model classes that don't extend ActiveRecord classes. I've used this approach to handle data that was stored in files rather than an RDBMS.
CRUD is an acronym for Create, Read, Update, Delete. Given a suitably-named database table, running a Rails scaffolding script will generate basic CRUD functionality for that table: a web page listing all the records in it (with paging logic if there are many) with options to edit or delete the existing records, or to create a new one.
I used VB a few years ago and was thrilled with it for 3 days, until I had to do some string manipulation. Rails can be similar, but note that the issue is at the framework level, not the language level. Ruby is a really great language. Rails is also a great framework, but you're more likely to run into cases with it where it doesn't work exactly how you'd like. In my experience, ActiveRecord (the database interface) is the most likely area to cause problems. In particular, it makes many assumptions about your data model that aren't likely to hold if you're not creating a schema from scratch. All of its assumptions are good relational practices, but that won't help if you need to work with a legacy schema. In this case, you could spend a lot of time trying to get ActiveRecord to work with your tables, or you could save yourself some trouble and grab the Ruby bindings for your database and take matters into your own hands.
Wow, thanks for the link! It never occurred to me that some of these gotchas were design decisions based on the lack of transactions:
TRADITIONAL
Make MySQL behave like a "traditional" SQL database system. A simple description of this mode is "give an error instead of a warning" when inserting an incorrect value into a column. Note: The INSERT/UPDATE aborts as soon as the error is noticed. This may not be what you want if you are using a non-transactional storage engine, because data changes made prior to the error are not be rolled back, resulting in a "partially done" update. (New in MySQL 5.0.2)
Of course, as the MySQL line has always been to manage transactions in code, I don't see why they didn't just raise an error and let the application deal with the manual rollback. Who would ever want invalid data to be inserted?
I second the vote for Dr. Mario. I've been seriously addicted to that game (Gameboy version) for years. It's downside is that it has a big twitch and luck factor at level 20+ at hi speed, though.
What's illegal about using P2P networks to share information you have a right to distribute? Always keep in mind that P2P isn't illegal, but sharing copyrighted material without permission to distribute it is.
They're actually right to be concerned. That's exactly why biometrics aren't well suited to many applications (particularly authentication). They are neither secret nor expirable, and can be stolen very easily. Passwords and PIN numbers will be with us for a very long time.
I shy from dos2unix because it's not very portable between *nixes. It's good to memorize the Perl way to do this. The parent actually missed the useful "-i" flag:
find lcc/src -type f | xargs perl -i -pe 's/\r//g'
You don't want a spider crawling over a webapp and creating, deleting, updating data it knows nothing about. If there's a new wave of robots.txt ignoring spiders, there's going to be a lot of ugly side effects.
The point of the article is that intelligent people are really good at defending bad ideas. So much so, in fact, that they can often convince other smart people that they are actually good ideas. This is the problem that needs to be addressed. The CEO of a startup I worked at for 5 years was exactly like this. He could come up with ideas and arguments at a mile a second and refused to lose an argument. The problem was that his ultra-fast ideas weren't always good, and I'd realize the holes in his arguments several hours later. The problem is that sometimes we had already made a firm decision based on the CEO's flawed assumptions.
This is one of the best articles I've read in a long time. People like this can literally ruin a company, and it's good to have a concrete strategy for combatting them because they are very formidable opponents in an argument.
The problem is that they won't get equal functionality in an open source VM, either. Writing a good JIT is a lot of work, and it needs to be done for every platform. Obscure platforms just won't get the same level of attention. It will probably be better than Sun's support, but still only the most popular platforms will get top-notch support. Unfortunately, it's just the way it works. Look at how frequently Linus breaks the compile for non-x86 platforms!
Re:8 bit propritary code ... hm ...
on
Archon to be Revived
·
· Score: 2, Insightful
There is always xarchon. Personally, I prefer to use VICE to emulate the original C64 game.
Java cannot run everywhere. Java can only run on platforms that have a useable JVM. If your only goal is to write an application to run "cross-distro" (which I take to mean Linux) that you might as well write something that can run on all flavors of UNIX. Pick Perl, Python, Ruby, or whatever blows your hair back. You'll find it much simpler than trying to get Java running on OpenBSD!
If you want something to work on Windows, Mac, and a subset of UNIX including Linux, Solaris, AIX, and HPUX, then Java is a reasonable choice. Using Java strictly for UNIX is actually counter-productive.
thank god I'm not the only one who's used crisco! every time I mention it I get some strange looks, but it's the best! i've even been nicknamed the "crisco cowboy", which i'm somewhat proud of, for some reason or other. we really do have no shame. at least I don't.
The bytecode for a method declares the maximum size the stack can grow to within that method. If this number is small enough, everything can be stored in registers.
That's exactly what I noticed. MS isn't trying to discredit Linux with FUD anymore because they seem to have learned that FUD is ineffective against OSS. Interestingly, the Halloween document highlighted this very point, but they only now are trying other tactics.
Notice how Taylor still tried to belittle Linux, though. When asked, "Where do you see Linux being successful today?" he immediately replied, "Definitely on the edge." He then further qualified it with, "You're just seeing edge services continue, such as firewall, appliances and those types of devices."
It's pretty clear that he's trying to marginalize Linux by characterizing as an "edge" operating system, but with a less hostile tone, as you pointed out. He neglects to mention the large installed base of Linux webservers, databases, mail servers, etc. He even implies that there aren't any other areas Linux is successful in when he's uses the word "just" in "You're just seeing edge services..." When faced with a direct question about Linux on the desktop, he only mentioned increasing "conversations" about it, but not the actual conversions which are taking place.
The other way he attempts to discredit Linux is to note that there is very little ISV support for it. While it's a true statement, it's quite misleading. I've been using Linux for many years without needing software from ISVs. Nearly every piece of software needed by the average user is available as Open Source! Taylor knows that much of the industry still believes ISV support is crucial to success.
I'm actually very impressed with Taylor. He didn't outright lie about anything, but he definitely painted Linux in an unfair light. However, he admitted to shortcomings in the current MS offerings, which makes him seem quite believable. I think this article could be quite persuasive to those who don't know much about Linux. That's probably a bad thing, but we're going to have to show a level of tact on par with Taylor's to sway people back toward OSS.
The author of the article has a very nice grasp of what Kermit is. It's not just a protocol, but a program complete with scripting capabilities, modem dialing, transfers using several protocols (including Kermit of course). It can even do TCP transfers now. It's a great program, but it's a little hard to use and mostly surpassed by simpler tools now. Still, I needed to use it a few years ago to automate modem uploads to a mainframe.
Have you ever tried VMware? It only emulates a few things, but for the most part apps run at native speed. I think this is the sort of answer the person was looking for, not "just install it on it's own partition." It's much easier to do than resize partitions or install a new drive. Granted, neither of these are hard to do, but it's an awful lot of trouble to just evaluate something, especially if you're not really planning on using it.
Slashdot Mirror
I should probably clarify that "taking matters into your own hands" doesn't mean you have to ditch Rails. You'll just write model classes that don't extend ActiveRecord classes. I've used this approach to handle data that was stored in files rather than an RDBMS.
CRUD is an acronym for Create, Read, Update, Delete. Given a suitably-named database table, running a Rails scaffolding script will generate basic CRUD functionality for that table: a web page listing all the records in it (with paging logic if there are many) with options to edit or delete the existing records, or to create a new one.
I used VB a few years ago and was thrilled with it for 3 days, until I had to do some string manipulation. Rails can be similar, but note that the issue is at the framework level, not the language level. Ruby is a really great language. Rails is also a great framework, but you're more likely to run into cases with it where it doesn't work exactly how you'd like. In my experience, ActiveRecord (the database interface) is the most likely area to cause problems. In particular, it makes many assumptions about your data model that aren't likely to hold if you're not creating a schema from scratch. All of its assumptions are good relational practices, but that won't help if you need to work with a legacy schema. In this case, you could spend a lot of time trying to get ActiveRecord to work with your tables, or you could save yourself some trouble and grab the Ruby bindings for your database and take matters into your own hands.
Of course, as the MySQL line has always been to manage transactions in code, I don't see why they didn't just raise an error and let the application deal with the manual rollback. Who would ever want invalid data to be inserted?
If it's truly anonymous, then you wouldn't be able to tell which one you're pointing at, because they'd have to be too close together!
I second the vote for Dr. Mario. I've been seriously addicted to that game (Gameboy version) for years. It's downside is that it has a big twitch and luck factor at level 20+ at hi speed, though.
What's illegal about using P2P networks to share information you have a right to distribute? Always keep in mind that P2P isn't illegal, but sharing copyrighted material without permission to distribute it is.
They're actually right to be concerned. That's exactly why biometrics aren't well suited to many applications (particularly authentication). They are neither secret nor expirable, and can be stolen very easily. Passwords and PIN numbers will be with us for a very long time.
I shy from dos2unix because it's not very portable between *nixes. It's good to memorize the Perl way to do this. The parent actually missed the useful "-i" flag:
find lcc/src -type f | xargs perl -i -pe 's/\r//g'
There's no need to exec dos2unix for each file individually. xargs is preferred for all (non-broken) apps that can take multiple file arguments.
And how do you assemble and link your code? The virus doesn't have to be injected during the translation from C to assembly...
Are you kidding?
u32 vote32(u32 x, u32 y, u32 z) {
return (x & y) | (y & z) | (x & z);
}
You don't want a spider crawling over a webapp and creating, deleting, updating data it knows nothing about. If there's a new wave of robots.txt ignoring spiders, there's going to be a lot of ugly side effects.
The point of the article is that intelligent people are really good at defending bad ideas. So much so, in fact, that they can often convince other smart people that they are actually good ideas. This is the problem that needs to be addressed. The CEO of a startup I worked at for 5 years was exactly like this. He could come up with ideas and arguments at a mile a second and refused to lose an argument. The problem was that his ultra-fast ideas weren't always good, and I'd realize the holes in his arguments several hours later. The problem is that sometimes we had already made a firm decision based on the CEO's flawed assumptions.
This is one of the best articles I've read in a long time. People like this can literally ruin a company, and it's good to have a concrete strategy for combatting them because they are very formidable opponents in an argument.
You missed the ever important:
bj > Fellatitations!
Yeah, I was thinking they should change the spelling to be more Apple compliant. Something like Rape-iScan.
The problem is that they won't get equal functionality in an open source VM, either. Writing a good JIT is a lot of work, and it needs to be done for every platform. Obscure platforms just won't get the same level of attention. It will probably be better than Sun's support, but still only the most popular platforms will get top-notch support. Unfortunately, it's just the way it works. Look at how frequently Linus breaks the compile for non-x86 platforms!
There is always xarchon. Personally, I prefer to use VICE to emulate the original C64 game.
Java cannot run everywhere. Java can only run on platforms that have a useable JVM. If your only goal is to write an application to run "cross-distro" (which I take to mean Linux) that you might as well write something that can run on all flavors of UNIX. Pick Perl, Python, Ruby, or whatever blows your hair back. You'll find it much simpler than trying to get Java running on OpenBSD!
If you want something to work on Windows, Mac, and a subset of UNIX including Linux, Solaris, AIX, and HPUX, then Java is a reasonable choice. Using Java strictly for UNIX is actually counter-productive.
thank god I'm not the only one who's used crisco! every time I mention it I get some strange looks, but it's the best! i've even been nicknamed the "crisco cowboy", which i'm somewhat proud of, for some reason or other. we really do have no shame. at least I don't.
The bytecode for a method declares the maximum size the stack can grow to within that method. If this number is small enough, everything can be stored in registers.
That's exactly what I noticed. MS isn't trying to discredit Linux with FUD anymore because they seem to have learned that FUD is ineffective against OSS. Interestingly, the Halloween document highlighted this very point, but they only now are trying other tactics.
Notice how Taylor still tried to belittle Linux, though. When asked, "Where do you see Linux being successful today?" he immediately replied, "Definitely on the edge." He then further qualified it with, "You're just seeing edge services continue, such as firewall, appliances and those types of devices."
It's pretty clear that he's trying to marginalize Linux by characterizing as an "edge" operating system, but with a less hostile tone, as you pointed out. He neglects to mention the large installed base of Linux webservers, databases, mail servers, etc. He even implies that there aren't any other areas Linux is successful in when he's uses the word "just" in "You're just seeing edge services..." When faced with a direct question about Linux on the desktop, he only mentioned increasing "conversations" about it, but not the actual conversions which are taking place.
The other way he attempts to discredit Linux is to note that there is very little ISV support for it. While it's a true statement, it's quite misleading. I've been using Linux for many years without needing software from ISVs. Nearly every piece of software needed by the average user is available as Open Source! Taylor knows that much of the industry still believes ISV support is crucial to success.
I'm actually very impressed with Taylor. He didn't outright lie about anything, but he definitely painted Linux in an unfair light. However, he admitted to shortcomings in the current MS offerings, which makes him seem quite believable. I think this article could be quite persuasive to those who don't know much about Linux. That's probably a bad thing, but we're going to have to show a level of tact on par with Taylor's to sway people back toward OSS.
The author of the article has a very nice grasp of what Kermit is. It's not just a protocol, but a program complete with scripting capabilities, modem dialing, transfers using several protocols (including Kermit of course). It can even do TCP transfers now. It's a great program, but it's a little hard to use and mostly surpassed by simpler tools now. Still, I needed to use it a few years ago to automate modem uploads to a mainframe.
The more common synonym is autocracy.
Have you ever tried VMware? It only emulates a few things, but for the most part apps run at native speed. I think this is the sort of answer the person was looking for, not "just install it on it's own partition." It's much easier to do than resize partitions or install a new drive. Granted, neither of these are hard to do, but it's an awful lot of trouble to just evaluate something, especially if you're not really planning on using it.
Standard xmodem uses a 128 byte block size. There is an extension called xmodem-1k which uses a 1024 byte block size.