Yes, but Oil companies are booming. Have a little heart for those Oil execs. They have feelings, too. Just ask anyone at the country club or the yacht club. Sheesh, for a while, even some countries were making more money than the oil companies. Thank god that's been rectified.
This demonstrates why it is important to document any "clever tricks" used in a software system.
At the time, it may not have seemed like a clever trick, just a logical way to do things given the tools at hand. I've looked at code I wrote a few years ago and had to think about my mindset at the time. We do document the code I think needs explanation but we don't document the obvious, like "balance += deposit". But, what seems obvious at the time may not seem obvious later.
The clever trick the poster referred to may very well have been documented, but it still doesn't mean it's going to jump out at you as a Y2K problem when you review the code.
Actually, it was a problem. I worked for a year on a Y2K team. We fixed alot of bugs that would have had serious consequences. For example, some bank code logged transactions using a YYMMDD prefix to force the transactions into a certain order. In one case, deposits made would have been lost because they were not in the right order. Yes, it was bad code, but it was there.
You don't panic, but you act. IMO, Y2K is a huge success story. The problems, some of them big, really were there.
Those of us who wrote software for these machines just laughed and repeated the mantra, "Embedded systems programmers don't use COBOL."
True, but...
First of all, Cobol didn't force you to use 2-Digit dates and embedded systems at that time were more memory constrained than they are today so the temptation was high to save space where ever possible. For example, apparently there were devices used in the electrical grid to route power based on expected usage which took into consideration weekends and holidays. Some of these used a 6 digit date which rolled back to 1900 and thus didn't have the right date after 1999 and would have routed power incorrectly resulting in brownouts or wasted power.
I worked with a team offering Y2K services for customers of a large computer company. We fixed alot of code and it would have been a much bigger problem than it was if it hadn't been taken seriously.
One tiny example, which showed up before 1/1/2000 was with a large nationwide toy chain. If you used a credit card with an expiration date after 1/1/2000 every register in that store hung. We found alot of issues that would have had news-making consequences if nothing had been done.
The fact that it was a non-event was because alot was done to fix the code before hand.
One interesting side note is that we hired a bunch of retired Cobol programmers because alot of the code we fixed was in Cobol. Although I hadn't done alot of Cobol programming myself, it became very clear to me that Cobol was at that point and probably still is the most effective Business programming language. Pretty amazing considering its age.
t is not phenomenally ignorant question to ask the wife. If her man beats her, why not leave?
That's not what the metaphore was saying at all. The question was "Why can't we all get along."
But, since you mention Doc files...
Writer is cool, but it can't pass DOC files back and forth with Word without screwing things up.
Not quite true (actually not true at all). Writer can pass DOC files around. We do it all the time. Back to the getting along question, OO accomplishes this task quite well with open standards and shared files freely with Sypmonie and others using open standards. It's MS who doesn't want to play along. Their answer was to create a competing standard that even they cannot implement.
The photoshop comparison shows a different animal alltogether. Photoshop is an excellent tool... as long as you are running Windows. If I'm not running Windows then Photoshop has a serious disadvantage! I use GIMP because it works for what I do, it works on Linux and I like the price. Those are my criteria. I haven't used photoshop but I hear it's excellent. If someone want's to create a real competitor to Photoshop, I think they could and I don't have a problem with their market share because they compete fairly. Anyone creating a better image editing tool that Photoshop can compete. You can't say the same about MS Office because they MS also has an effective monopoly on the file format. And they know as well and you and I that to gain market share in the Office market, you have to be compatible with the files being passed around.
This is the hand you are dealt, you have to play hard to win.
Agreed, but let's start with being able to compete in the first place with open standards and a level playing field.
You've made a valid comparison regarding common morality vs. Microsoft policy
M$ trolls w/ mod points is the prob here
I am always amazed at the lack of understanding that most MS fans have about the reality of the software and OS market. They seem to accept the viruses and lack of choice as normal. A good friend whose a rabid MS fan thinks it's normal to have to reformat your harddrive every year to get rid of viruses. It takes him a weekend to do it by the time he gets all his software and drivers installed.
I did chuckle when he called me once to ask me if I could download some drivers for him. Apparently Windows didn't recognise his network card out of the box so he had to download a driver to get it work, which he couldn't do because he had no network connectivity. So, I downloaded driver for him on my Linux box. His only other option was to boot a Linux live CD to download the driver!
I don't understand the desire of many slashdotters to see For Profit software companies fail.
I think the issue is not that most slashdotters wish for For Profit software companies to fail. The issue is that MS is a monopoly and has a history of using every trick in the book to make sure noone threatens that monopoly. This goes way back to DOS 3 when internal memos revealed that they were manipulating DOS to make sure Lotus 123 didn't work well on it. "DOS isn't done until Lotus won't run".
Our development group was affected when they came out with Visual C++. We were heavily invested in Borland C++. Every OS release seemed to break Borland (i.e. it worked fine until we upgraded Windows). Remember when it came out that MS had created secret APIs for Windows which only their own developers knew about?
It's as if GM were to own all the roads in the US and design them so that only GM cars could run on them. MS owns the road and makes it extremely difficult for any competing product to work well on it.
I use Ext2 IFS [fs-driver.org] in Windows (it works for Ext3 too) and it is, hands-down, the stablest and best Ext2/3 Windows driver I've used.
Thanks for that tip. I just installed it. I haven't done much with it, but I am already pretty impressed. It installed quickly and works seemlessly. This solved a problem I've had for a while with an external drive I formatted with ext3 which I can now sync with my laptop.
I'm not a fan of IP -- it's like patenting math, and many trademarks are not far behind. But, Microsoft has used it's legal muscle and deep pockets to outlast others who have had "valid" (in the sense of legally valid in this system) trademarks and IP patents.
MS wants to have their cake and eat it, too. OTOH they patent things like using a password to get administrative authority (sounds like sudo to me), on the other hand, their SQLServer DB was so similar to Sybase that use could actually use the same drivers.
The reason this company waited is to get more money from MS. MS knows they went out on a limb on this.
Reminds me of a story about Wang Computers (this was a long time ago). There was no internet at that time, but Wang, like DEC, had an internal text-based blogging system where people could post notes of interest and others could respond to those posts.
Apparently, there was a post having to do with cars (probably muscle cars) which had thousands of responses. Someone noticed that there was another post about Sex with only a few responses over a similar time frame and started a post about this curious phenomon.
Finally someone figured it out. The reason there was so much more interest in the cars post was that Wang Engineers do have cars...
And besides, as a public figure he's a fair target for satire.
Agreed. Additionally, the site is very much about Stephen Conroy and his actions. IMO the auDA has completely overstepped their bounds. This is just plain censorship of the USSR vintage.
They just want to cost Vimeo money so the next guy or company who wants to do something perfectly legal will be afraid.
Can you imagine what the record companies would do in a world where artists marketed their works directly to the customer? There's only one thing to do: make that illegal or at least sue them out of existence.
I wasn't trying to imply that PG is better than everything else for all workloads. But, for relational database scenarios, I've found that PG is faster and more secure than MySQL. We've noticed a significant performance improvement even comparing PG to MyISAM tables. Given that,
Facebook, Twitter, Amazon, etc, I believe use Key-Value stores for performance reasons. Hadoop is a big player in that game, Google, Yahoo and IBM are using it. Not sure about Amazon, Facebook, Twitter and so on.
I moved our application of now over 1 millions lines from MySQL to Postgres with only a few minor changes and no stored procedures.
I'm not sure what you mean by "I can write SQL in a way that upgrades toward a situation instead of failing when the incorrect old situation was there." Can you elaborate?
In my experience, MySQL doesn't offer any more flexibility on the whole than Postgresql. Some stuff is better with MySQL, like replication, whereas other stuff is better with Postgres (performance of joins, variable length character fields, etc). But, mainly the reason for preferring PG over MySQL is performance.
Especially since MySQL is already well known for both data loss and corruption in the name of performance. Made all the more embarrassing is that PostgreSQL consistently either meets or beats MySQL in performance and leaves it far behind in scalability.
We moved our DB from MySQL to PG about a year ago and experienced significant performance improvements over both InnoDB and MyISAM. Pretty interesting when you consider that PG is also more secure that MySQL.
My guess is that once PG reaches critical mass as an Open Source DB name, people will be moving away from MySQL in groves.
Slashdot Mirror
Yes, but Oil companies are booming. Have a little heart for those Oil execs. They have feelings, too. Just ask anyone at the country club or the yacht club. Sheesh, for a while, even some countries were making more money than the oil companies. Thank god that's been rectified.
This demonstrates why it is important to document any "clever tricks" used in a software system.
At the time, it may not have seemed like a clever trick, just a logical way to do things given the tools at hand. I've looked at code I wrote a few years ago and had to think about my mindset at the time. We do document the code I think needs explanation but we don't document the obvious, like "balance += deposit". But, what seems obvious at the time may not seem obvious later.
The clever trick the poster referred to may very well have been documented, but it still doesn't mean it's going to jump out at you as a Y2K problem when you review the code.
an executive can't lose his house for big failures when the company goes bankrupt
I believe you can be sued for negligence, though.
Actually, it was a problem. I worked for a year on a Y2K team. We fixed alot of bugs that would have had serious consequences. For example, some bank code logged transactions using a YYMMDD prefix to force the transactions into a certain order. In one case, deposits made would have been lost because they were not in the right order. Yes, it was bad code, but it was there.
You don't panic, but you act. IMO, Y2K is a huge success story. The problems, some of them big, really were there.
Those of us who wrote software for these machines just laughed and repeated the mantra, "Embedded systems programmers don't use COBOL."
True, but ...
First of all, Cobol didn't force you to use 2-Digit dates and embedded systems at that time were more memory constrained than they are today so the temptation was high to save space where ever possible. For example, apparently there were devices used in the electrical grid to route power based on expected usage which took into consideration weekends and holidays. Some of these used a 6 digit date which rolled back to 1900 and thus didn't have the right date after 1999 and would have routed power incorrectly resulting in brownouts or wasted power.
I worked with a team offering Y2K services for customers of a large computer company. We fixed alot of code and it would have been a much bigger problem than it was if it hadn't been taken seriously.
One tiny example, which showed up before 1/1/2000 was with a large nationwide toy chain. If you used a credit card with an expiration date after 1/1/2000 every register in that store hung. We found alot of issues that would have had news-making consequences if nothing had been done.
The fact that it was a non-event was because alot was done to fix the code before hand.
One interesting side note is that we hired a bunch of retired Cobol programmers because alot of the code we fixed was in Cobol. Although I hadn't done alot of Cobol programming myself, it became very clear to me that Cobol was at that point and probably still is the most effective Business programming language. Pretty amazing considering its age.
t is not phenomenally ignorant question to ask the wife. If her man beats her, why not leave?
That's not what the metaphore was saying at all. The question was "Why can't we all get along."
But, since you mention Doc files...
Writer is cool, but it can't pass DOC files back and forth with Word without screwing things up.
Not quite true (actually not true at all). Writer can pass DOC files around. We do it all the time. Back to the getting along question, OO accomplishes this task quite well with open standards and shared files freely with Sypmonie and others using open standards. It's MS who doesn't want to play along. Their answer was to create a competing standard that even they cannot implement.
The photoshop comparison shows a different animal alltogether. Photoshop is an excellent tool ... as long as you are running Windows. If I'm not running Windows then Photoshop has a serious disadvantage! I use GIMP because it works for what I do, it works on Linux and I like the price. Those are my criteria. I haven't used photoshop but I hear it's excellent. If someone want's to create a real competitor to Photoshop, I think they could and I don't have a problem with their market share because they compete fairly. Anyone creating a better image editing tool that Photoshop can compete. You can't say the same about MS Office because they MS also has an effective monopoly on the file format. And they know as well and you and I that to gain market share in the Office market, you have to be compatible with the files being passed around.
This is the hand you are dealt, you have to play hard to win.
Agreed, but let's start with being able to compete in the first place with open standards and a level playing field.
"In my home, at work, and on the street; I tell people that m$ is like the scorpion on the turtles back."
I'm sure you have lots of friends...
There are alot of people who share his opinion.
You've made a valid comparison regarding common morality vs. Microsoft policy
M$ trolls w/ mod points is the prob here
I am always amazed at the lack of understanding that most MS fans have about the reality of the software and OS market. They seem to accept the viruses and lack of choice as normal. A good friend whose a rabid MS fan thinks it's normal to have to reformat your harddrive every year to get rid of viruses. It takes him a weekend to do it by the time he gets all his software and drivers installed.
I did chuckle when he called me once to ask me if I could download some drivers for him. Apparently Windows didn't recognise his network card out of the box so he had to download a driver to get it work, which he couldn't do because he had no network connectivity. So, I downloaded driver for him on my Linux box. His only other option was to boot a Linux live CD to download the driver!
Well put.
I don't understand the desire of many slashdotters to see For Profit software companies fail.
I think the issue is not that most slashdotters wish for For Profit software companies to fail. The issue is that MS is a monopoly and has a history of using every trick in the book to make sure noone threatens that monopoly. This goes way back to DOS 3 when internal memos revealed that they were manipulating DOS to make sure Lotus 123 didn't work well on it. "DOS isn't done until Lotus won't run".
Our development group was affected when they came out with Visual C++. We were heavily invested in Borland C++. Every OS release seemed to break Borland (i.e. it worked fine until we upgraded Windows). Remember when it came out that MS had created secret APIs for Windows which only their own developers knew about?
It's as if GM were to own all the roads in the US and design them so that only GM cars could run on them. MS owns the road and makes it extremely difficult for any competing product to work well on it.
It just keeps getting weirder and weirder.
I use Ext2 IFS [fs-driver.org] in Windows (it works for Ext3 too) and it is, hands-down, the stablest and best Ext2/3 Windows driver I've used.
Thanks for that tip. I just installed it. I haven't done much with it, but I am already pretty impressed. It installed quickly and works seemlessly. This solved a problem I've had for a while with an external drive I formatted with ext3 which I can now sync with my laptop.
Thanks!
Apparently they figured out that only Windows and Mac users still believe in Santa. The Linux users were just too damned smart.
Perhaps that's the reason why Wang failed: computers are like errections, if their up and you want to keep them up, don't screw with them.
As opposed to, say, Wangs?
Excellent!
Actually, I think the implication was the Wang Engineers have cars but not sex.
I'm not a fan of IP -- it's like patenting math, and many trademarks are not far behind. But, Microsoft has used it's legal muscle and deep pockets to outlast others who have had "valid" (in the sense of legally valid in this system) trademarks and IP patents.
MS wants to have their cake and eat it, too. OTOH they patent things like using a password to get administrative authority (sounds like sudo to me), on the other hand, their SQLServer DB was so similar to Sybase that use could actually use the same drivers.
The reason this company waited is to get more money from MS. MS knows they went out on a limb on this.
IMO, MS is reaping what they sow.
Bung sounds like what happens to you in prison: "Damn, Bubba bunged me again."
Reminds me of a story about Wang Computers (this was a long time ago). There was no internet at that time, but Wang, like DEC, had an internal text-based blogging system where people could post notes of interest and others could respond to those posts.
Apparently, there was a post having to do with cars (probably muscle cars) which had thousands of responses. Someone noticed that there was another post about Sex with only a few responses over a similar time frame and started a post about this curious phenomon.
Finally someone figured it out. The reason there was so much more interest in the cars post was that Wang Engineers do have cars...
And besides, as a public figure he's a fair target for satire.
Agreed. Additionally, the site is very much about Stephen Conroy and his actions. IMO the auDA has completely overstepped their bounds. This is just plain censorship of the USSR vintage.
They just want to cost Vimeo money so the next guy or company who wants to do something perfectly legal will be afraid.
Can you imagine what the record companies would do in a world where artists marketed their works directly to the customer? There's only one thing to do: make that illegal or at least sue them out of existence.
I wasn't trying to imply that PG is better than everything else for all workloads. But, for relational database scenarios, I've found that PG is faster and more secure than MySQL. We've noticed a significant performance improvement even comparing PG to MyISAM tables. Given that,
Facebook, Twitter, Amazon, etc, I believe use Key-Value stores for performance reasons. Hadoop is a big player in that game, Google, Yahoo and IBM are using it. Not sure about Amazon, Facebook, Twitter and so on.
I moved our application of now over 1 millions lines from MySQL to Postgres with only a few minor changes and no stored procedures.
I'm not sure what you mean by "I can write SQL in a way that upgrades toward a situation instead of failing when the incorrect old situation was there." Can you elaborate?
In my experience, MySQL doesn't offer any more flexibility on the whole than Postgresql. Some stuff is better with MySQL, like replication, whereas other stuff is better with Postgres (performance of joins, variable length character fields, etc). But, mainly the reason for preferring PG over MySQL is performance.
They aren't dumb, just expensive. If they cost more than their worth, you don't need them.
True, but PostgreSQL is faster and more secure. Why not have both if you can?
Especially since MySQL is already well known for both data loss and corruption in the name of performance. Made all the more embarrassing is that PostgreSQL consistently either meets or beats MySQL in performance and leaves it far behind in scalability.
We moved our DB from MySQL to PG about a year ago and experienced significant performance improvements over both InnoDB and MyISAM. Pretty interesting when you consider that PG is also more secure that MySQL.
My guess is that once PG reaches critical mass as an Open Source DB name, people will be moving away from MySQL in groves.