And you're making assumptions on the complexity of the code and whether those 4200 lines of Java actually did produce an acceptable alternative. As opposed to throwing good money, brainpower and time after bad and never producing anything acceptable at all? That benefits people how? When the project's a WOMBAT, it's time to ditch and start over.
On the bright side, that rewrite didn't make things any worse. Adding 4200 lines of [original obscure language] to the Big Ball of Mud wouldn't have had even that good an effect...
Until you win, then the other side gets to pick up the bill. Not always. It's technically up to the court to decide whether costs are awarded. Yes, that's normally done (as it discourages time-wasting and frivolous suits) but the courts can and will refuse the winner the costs when they think to do otherwise would be unjust or an abuse of process. For an example, see the McLibel case (well, before the appeals are taken into account; they make everything more complex and hammer the point home harder).
Companies with full data centers and in need of more servers are turning to virtualization technologies to increase their server density, reduce their physical server deployment, and improve efficiency in cooling, hardware maintenance, and administration. It buys you a few years, but that's all. There's massive growth going on in the amount of server capacity needed, and all virtualization can do is to get your utilization up closer to the 95% level (bad idea to go much over that; you need a little space for admin overhead). But once you've done that, you're going to need more real capacity anyway (or the business isn't growing...) At best, use virtualization to buy yourself the time to get your physical server systems in order so you can host more physical kit. And do consider out-sourcing for at least some functions; for example, Amazon EC2 is cheap.
You know in England when they switched to metric in bars they went from getting a pints to half litres and the price went up, man. Beer is still sold in pints (and half-pints) in England. Really.
Even if they were twisted out of 6 foot tall blond virgins' pubic hair, it's still not worth it! Well, we've got virgins galore on/. and some of them must be 6 foot tall and blond, so I think a live test should be possible...
Joking aside I'd really like to know how this dramatic change came about. He was replaced by a robot from Neocon Central Command as soon as it looked like he might actually win the nomination. The real McCain is probably being "entertained" in some dark cellar in deepest darkest Utah...
Corporations don't do as we want, but they do listen when we hit their wallet. This only works in a national economy, which we no longer have. The corporations are multinational, and have six billion prospective customers. Your purchase is meaningless and there is no way possible to hit them in their wallet. They have no reason to care if you buy or not, there are a lot more suckers where you came from. That bullshit is what they want you to believe, since it encourages feelings of helplessness and acts to prevent most people from joining a boycott (which would hurt, as corporations tend to have substantial costs as well and the boycott doesn't cut those). Instead, you absolutely should not patronize any corporation that does stuff you're against, and better yet, tell other people that you're doing this too. Use the power of the internet to help people find out the facts (stick to those though, please; no need to punish anyone for pure hearsay and rumor) and learn to persuade others that you're not only right but worth joining.
For example, I don't like the business practices of the "music business" at all so I won't buy their products. I also won't pirate them (I'm law-abiding) but I've got plenty of other things to do (writing OSS) that I feel I don't miss anything much. But I will and do support live music; I love going to concerts and the like. (I also mostly avoid cinema, but that's because most films are a load of fetid whale dreck. If the studios want my money, they have to produce something worth it first; I don't mind paying for stuff that's good.)
Does it use the same number of pixels per channel? I hope not. Here's why: the human eye is not equally sensitive to each of the three primary colors; we can see quite a lot finer differences in green than in blue (red comes between the two extremes). To show this, create a simple monochromatic stepped gradient image in green and another in blue. Now eyeball them using a viewer that doesn't do fancy gamma correction; on a 24bpp display you should be able to see the steps on the green image (assuming normal color vision) but you'll have real problems doing that with the blue image.
As much as I think TPTB would like to kill off truecrypt Hmm. Perhaps I should get more sleep, since I misread that as being The Pirate Bay wanting to shut down truecrypt and wondered what on earth for. It took a shameful amount of time to realize what you were actually saying...
From using a record company as your ISP. They're almost completely unrelated companies; the only thing in common (apart from some shared shareholders) is the fact that they both license the "Virgin" trademark from the same third company.
If I buy a car from a rental company I can do with it what I wish within the bounds of the law (and optionally physics.) I think you'll find that the dictates of the laws of physics are enforced rather more stringently than any human laws (except perhaps those relating to taxation of ordinary people).
This talk was one of the highlights of the conference. At the talk, they showed performance benchmarks that included running several things as much as 117x as fast as the default Ruby interpreter that is in use by most Rails installations today. The fact that it's built on this commercial-grade Gemstone platform that has been used for years for high-performance production Smalltalk applications just adds to its credibility. Well, without in any way detracting from the technical prowess of the Gemstone folks - sounds like they know their stuff with OODBs - beating the default implementation Ruby on the performance front isn't that hard. Certainly traditionally, Ruby was slow by all independent measures such as the Language Shootout.
I suppose this does beg the questions: does the Gemstone-based system do better in the shootout benchmarks? If so, by how much? And since there's often tradeoffs between optimizations and (frequently obscure) parts of language semantics, what features of Ruby get lost along the way? What proportion of Ruby scripts are likely to run unchanged? What proportion of those are going to get a good performance boost?
Just the things that go through my head on first glance...
2)You don't get peanuts or anything else, for that matter, anymore, unless you're prepared to pay for them. Not on long-haul. Mind you, you pay for in the ticket price too.
I drive whenever I can, now. You drive intercontinental? That's... extreme.
I'll stick to Smalltalk because it's a good example for this discussion. OK. How many production systems are out there being run by Smalltalk code? (I'll exclude developer tools from this discussion; they're often written in the language they support.) The real mark of a successful language is that it is out there, supporting people doing real work, and it is holding its own in that particular area because it does the job well.
Now, I've never heard of a production system that runs on Smalltalk, but that doesn't mean such systems don't exist. Just nobody talks about them. (By comparison, I know for sure that there have been production systems running on Lisp in the past, even though I'm not sure how many are left these days.)
Say, isn't that little yellow lock supposed to mean something? Oh that's right. It means you have an SSL session with a web site that has a certificate issued from a trusted certificate authority like Verisign. Hmmmm.
So what is the difference again? At the cheaper end of CAs (*ahem*Godaddy*ahem*) about all they check is that you are the domain owner, which is about the absolute minimum to make SSL work correctly. What they don't check is whether the domain is owned by someone with an identity worth a damn; it's that check which is expensive and time-consuming. And the value to users? If the site scams you, you can reasonably easily track down who (in a corporate/legal sense) scammed you and sue them to oblivion. That is, you've got a guarantee of the corporation's good name, which is actually a reasonable brand most of the time, especially if the want to keep on trading.
Most people[*] don't need that level of assurance for personal certificates, and a lot of websites don't need it for their host certificates either. But for a home banking website, it's quite nice to have an additional layer that makes things harder for the bad guys to penetrate. (Not that this is the only precaution that should be taken, of course.)
Not all CAs are the same, and nor should you trust all certificates the same. The only thing at all reliable about them is the assertion "this is the certificate for site such-and-such", and the statement of who signed the certificate. And even those are only reliable if the CA certificate is one you trust, directly or otherwise...
[* I've come across applications which did need this level, but in that case the CAs tended to be paranoid people you had to go to in person and show multiple forms of id to. Which I suppose is perilously close to being security theatre in itself... ]
Now if we had done something really COOL, like drive there in a Jeep Commander, we would have used 22,263,157 gallons of gas and been MUCH better prepared for Mars. Let me know when you've put in a surface to drive on all the way to Mars. Doesn't have to be top-grade tarmac though.
It's a big deal because they have to get up and go to court, instead of sitting there and clicking their WHOIS button.
And when they'll get back all the good donuts will be gone! They should learn to think more positively! On the way back from the courthouse, they can stop off at Dunkin' Donuts and get another batch. Two birds, one stone.
I mean; when people to queue up to buy something when they're not even sure what it IS that's a whole new level of success. Except it's not new. I'm sure it's been done before in other sales areas, and probably on a regular basis.
But with only one button, they couldn't do much. That's OK. Since they'd be Apple iTanks, they'd have multitouch and gesture support. You stroke them one way and they crush the hordes of Redmond, you stroke them another and they blow you up for not having a cool enough fashion designer for a SO. And you thank them for it!
Like the ole Timex watch that "took a licking and kept on ticking" my desktop box, an ancient AMD Sempron 2600+ with a VIA chipset, unknown to me, lost its power connector to the CPU fan, which I only discovered by accident when replacing a hard disk drive. The CPU was hot enough to scald my finger, but neither its performance nor its stability has suffered one bit. It was quite possibly running faster because of the fan failure. This is because higher temperatures increase the charge carrier mobility in the silicon; the device is literally faster (I've seen this demonstrated for real on clockless hardware; impressive how large a speed variation was supported on what was essentially normal CMOS VLSI from about 7 years ago). On the other hand, I don't know whether the increased speed can lead to yet more heat and hence to more speed in a runaway reaction; it'd be fun if it could...
Which is why most packaging systems let you specify multiple repositories... Sure, but then you've got to support each of the different packaging and updating systems that each platform has invented, QAing the update process on each. It's a PITA, especially when you're trying to deal with platforms where there isn't even a commitment to a stable ABI of critical libraries between patchlevel releases...
I think I'd rather try to forget the pain than relive it.
Bollocs, the former code is completely immune to injection attacks because it's not accepting any external input (the sql sentence is a *constant*.) You write SQL code where everything you do in relation to it is always constant? I admire your industry, but respectfully think you're an idiot.
The problem wasn't the example per se. The problem was that it didn't promote good practices to a noob. That's cruel, and not just to the noob either. I hate being drafted in to clean up after someone else's mess that they made through cluelessness, and I bet you do too. Consistently teach people to do stuff the right way, and when it comes to somewhere where it matters, they'll get it right and the rest of us won't catch it in the neck down the line. Self-preservation really.:-)
Explain to me how you can 'inject' anything into a string literal. Nowhere in the post does he build a SQL statement from user input. Methinks you are just a little too ready to bust someone. The problem isn't the code example as written, but you wouldn't use that code as written (well, not more than once). The natural ("naive/clueless") extension is to use string substitutions to build the SQL string and that's where you start to get into trouble. Sure, for a little private database that's no big deal either, but such code practices tend to spread into places (e.g. webpage back ends). Like bubonic plague or Microsoft Exchange...
When writing examples, do it right. You're doing the questioner and the community a service.
1) My example was done to show how quick and easy it would be to set up and populate a table in Python.
2) He's explicitly mentioned several times that this is for his own private non-web use. But when doing so using good practices is so amazingly easy, why do even demo examples in a different way? It's not like you're using Java or C#; the right way doesn't need to be bureaucratic...
BTW, I checked the python docs and I'm happy to report that they strongly promote doing it the right way too. Thus you can do it right for almost the same amount of code:
mydb.execute("insert into contacts values(?,?,?)", ("Spooky", "Monster", "spook@spammity.spam"))
(I could comment on the merits of named versus positional parameters, but that's getting too close to language flaming. Off topic for here, and uninformative in any case.)
3) Your fly is open. Now you're just making yourself look like a silly kid.
Slashdot Mirror
On the bright side, that rewrite didn't make things any worse. Adding 4200 lines of [original obscure language] to the Big Ball of Mud wouldn't have had even that good an effect...
For example, I don't like the business practices of the "music business" at all so I won't buy their products. I also won't pirate them (I'm law-abiding) but I've got plenty of other things to do (writing OSS) that I feel I don't miss anything much. But I will and do support live music; I love going to concerts and the like. (I also mostly avoid cinema, but that's because most films are a load of fetid whale dreck. If the studios want my money, they have to produce something worth it first; I don't mind paying for stuff that's good.)
I'm sure you can come up with other examples.
Does it use the same number of pixels per channel? I hope not. Here's why: the human eye is not equally sensitive to each of the three primary colors; we can see quite a lot finer differences in green than in blue (red comes between the two extremes). To show this, create a simple monochromatic stepped gradient image in green and another in blue. Now eyeball them using a viewer that doesn't do fancy gamma correction; on a 24bpp display you should be able to see the steps on the green image (assuming normal color vision) but you'll have real problems doing that with the blue image.
I suppose this does beg the questions: does the Gemstone-based system do better in the shootout benchmarks? If so, by how much? And since there's often tradeoffs between optimizations and (frequently obscure) parts of language semantics, what features of Ruby get lost along the way? What proportion of Ruby scripts are likely to run unchanged? What proportion of those are going to get a good performance boost?
Just the things that go through my head on first glance...
Now, I've never heard of a production system that runs on Smalltalk, but that doesn't mean such systems don't exist. Just nobody talks about them. (By comparison, I know for sure that there have been production systems running on Lisp in the past, even though I'm not sure how many are left these days.)
So what is the difference again? At the cheaper end of CAs (*ahem*Godaddy*ahem*) about all they check is that you are the domain owner, which is about the absolute minimum to make SSL work correctly. What they don't check is whether the domain is owned by someone with an identity worth a damn; it's that check which is expensive and time-consuming. And the value to users? If the site scams you, you can reasonably easily track down who (in a corporate/legal sense) scammed you and sue them to oblivion. That is, you've got a guarantee of the corporation's good name, which is actually a reasonable brand most of the time, especially if the want to keep on trading.
Most people[*] don't need that level of assurance for personal certificates, and a lot of websites don't need it for their host certificates either. But for a home banking website, it's quite nice to have an additional layer that makes things harder for the bad guys to penetrate. (Not that this is the only precaution that should be taken, of course.)
Not all CAs are the same, and nor should you trust all certificates the same. The only thing at all reliable about them is the assertion "this is the certificate for site such-and-such", and the statement of who signed the certificate. And even those are only reliable if the CA certificate is one you trust, directly or otherwise...
[* I've come across applications which did need this level, but in that case the CAs tended to be paranoid people you had to go to in person and show multiple forms of id to. Which I suppose is perilously close to being security theatre in itself... ]
And when they'll get back all the good donuts will be gone! They should learn to think more positively! On the way back from the courthouse, they can stop off at Dunkin' Donuts and get another batch. Two birds, one stone.
I think I'd rather try to forget the pain than relive it.
The problem wasn't the example per se. The problem was that it didn't promote good practices to a noob. That's cruel, and not just to the noob either. I hate being drafted in to clean up after someone else's mess that they made through cluelessness, and I bet you do too. Consistently teach people to do stuff the right way, and when it comes to somewhere where it matters, they'll get it right and the rest of us won't catch it in the neck down the line. Self-preservation really.
When writing examples, do it right. You're doing the questioner and the community a service.
2) He's explicitly mentioned several times that this is for his own private non-web use. But when doing so using good practices is so amazingly easy, why do even demo examples in a different way? It's not like you're using Java or C#; the right way doesn't need to be bureaucratic...
BTW, I checked the python docs and I'm happy to report that they strongly promote doing it the right way too. Thus you can do it right for almost the same amount of code:
mydb.execute("insert into contacts values(?,?,?)", ("Spooky", "Monster", "spook@spammity.spam"))
(I could comment on the merits of named versus positional parameters, but that's getting too close to language flaming. Off topic for here, and uninformative in any case.) 3) Your fly is open. Now you're just making yourself look like a silly kid.