I don't care how good a computer scientist you are, the code you write in the first year or two you use a particular language is crap.
Sure, you understand the fundamentals of computer languages like flow control and logical operators, but you don't have a practical understanding of how to best use features of a particular language. A couple days with a book lets you write programs that you look back on a few years later and wonder "What the hell was I thinking when I wrote that?"
I have yet to meet a computer scientist right out of college that truly understands the tools used in information systems. They may or may not know how to write programs in languages used in the business world. They might have a limited grasp of relational databases but they've probably never encountered multi-dimensional databases. They certainly never bothered with anything as antiquated as mainframes.
It takes years of experience to become a "real" computer scientist instead of just another bozo with a sheepskin.
A properly trained professional with advanced experience will write a better product. I agree completely.
The problem is that not only do you confuse education with traiing and experience, you misunderstand the nature of computer science. Computer scientists learn about things like operating system design, language theory, hardware architecture and artificial intelligence. None of those things are applicable to writing business applications.
For business applications you're better off hiring someone with an information systems background. They've learned about how computers are used in the real world and have probably done an internship where they've had to work with business applications.
You missed the point. Some of the time, you don't need super speed as much as you need inexpensive to maintain.
Oh, and you should be using C instead of C++. Sure the overhead your talking about isn't much but why waste the hardware on overhead when you can take full advantage of it.
You don't need to reimplement COM and OLE just to parse the file. At most you might have to reimplement part of the Storage API to get to the streams. And you'd only have to do that if you were A) targeting a non-Windows platform and B) didn't want to grab some existing code from an open-source project like Calc.
You might need COM and OLE to display the file as it would appear in Office, but not to parse it.
Smaller tags = smaller file = faster reads. Smaller tags = fewer operations required to distinguish tags = faster parsing. The speed of string comparisons is always determined by the length of the strings. Checking if a tag matches 'somereallylongdescriptivetagname' will always take longer than comparing against 's'. There are some optimizations you can do based on the hardware. You can compare 4 case sensitive letters at a time on 32-bit hardware or 8 case sensitive letters at a time on 64-bit hardware. For case insensitive comparisons like xml, you're pretty much stuck with letter at a time.
As for why CSV isn't satisfactory for spreadsheets, make friends with an accountant or financial analyst and get them to show you how they use formulas, what-if scenarios, pivot-tables, and all of the other features of a modern spreadsheet program. CSV with a statistical analysis program is great for scientific data, but it doesn't cut it for financials.
I won't comment on the "collaboration" features, but a big part of Sharepoint and Domino are document repositories. Both of them have some major drawbacks in that area.
The big end user problem in Domino is the limited hierarchy. You have rooms, cabinets, folders, and files. People find it very limiting and confusing to be forced into that model. Some other issues involve poor performance and difficulty of performing backups because of certain design decisions that haven't been changed in the last 10 years. Everyone I've spoken to in my company hates using Domino.
Sharepoint offers unlimited hierarchy. The big problem in Sharepoint is security. You can set security on a respository but not on folders or documents. As far as I can determine, Windows authentication is required. This can be a real problem in a large corporation where various parts of the business have their own domains or active directory trees that aren't configured to trust the other domains or directories. Also, documents are differentiated and versioned entirely based on filename.
Sure, there are some things about document management that can be hard. I know from experience. I've written a document management system. That's why it amazes me that IBM and Microsoft haven't been able to put forth better offerings.
Interesting that your biggest problem is IE. My biggest problem is the lack of modal windows in Firefox and earlier versions of Safari. IE is the corporate standard, but employees working from home over VPN sometimes use other browsers. Many intranet applications require the use of modal dialogs and half-ass attempts to simulate modality do not make the cut.
IT wants to switch to Firefox but can't because Firefox doesn't provide a way to do a true modal dialog. There seems to be some kind of religious belief that modal windows are Evil but modals from alert(), confirm(), and prompt() are ok. I don't understand the theological difference between the two, but what I do understand is that the method Firefox's evangelism team advocates instead of showModalDialog does not offer the same functionality and tends to increase development and QA time.
We give project stakeholders the choice up front. We present their options in an unbiased fashion and let them decide. On every single project, the choice has been to use showModalDialog because it enables a consistent user interface without trying to anticipate and block all the different ways to navigate in the parent window.
Just about everything else can be made to work in every browser without much work. If the Firefox developers would follow in Safari's footsteps and implement showModalDialog, I could write applications that work in every major browser without compromising functionality or appearance. Then I might be complaining about the lack of CSS attributes like min-width in IE6. Unfortunately I think IE is going to catch up on the rendering side before Firefox implements anything because of business needs.
Clearly you don't work for a large organization. Applying a critical patch within week after it becomes available simply does not happen. Within a month, maybe. More likely 2-3 months. A regular patch takes 1-2 years. This is because the patches have to be tested against various configurations before being dissected and repackaged. Then the repakaged patch has to be tested against various configurations. The repackaging and testing occurs several times because someone always decides that they know better than the vendor which bits are actually required and leaves something out which later on turns out to be important.
We've been hit by a virus because some senior executive in one of the Manhattan offices decided to ignore policy and not put his internet web servers in the DMZ. Unfortunately, he's too powerful to get more than a slap on wrist. After the first attack I configured my servers to not accept any traffic from that block of addresses. I didn't patch my servers for about 3 months because no one is allowed to install anything that hasn't been repackaged and added to the standard operating environment. Senior management can ignore their own policies, but heaven help anyone else.
Sound stupid? It is. But it's par for the course in any really large organization.
Most of my consumer electronics are Sony. Receiver, DVD, TV, Playstation, PS2, etc. I have a rack full of PS and PS2 games and accessories. When I bought my PS2 on launch day, I thought $300 was a bit pricey. I could justify it to myself because stand alone DVD players were similar in price. I already had a DVD player but it would be nice to have one for the second TV as well. DVD offered clearly better picture quality than VHS on the TVs I already owned.
I have zero interest in a Blu-Ray or HD-DVD stand-alone player. I don't have an HDTV because HD programs only started being shown in my area two years ago. Even now, the selection of HD programming is very slim. My TVs probably have at least 5 years left on them before I start thinking about replacing them. Now, I'd be interested in a Blu-Ray or HD-DVD drive for my computer, but that's a different story.
So, I don't have HDTV. My existing DVD player outputs higher resolution than my TV can display. I have no desire for a Blu-Ray player. Why am I going to spend $500-$750 on a PS3? I'm not. I don't care how many killer games get released as PS3 exclusives. Until the price is $300 or less, I won't even consider it. Even then I still might not since I'm not happy about DRM.
If the PS3 costs substantially more than the XBox, Sony loses this generation of the consoles. It gets too expensive and Nintendo might even make it back into 2nd place.
Ok. PASS the car on the right and then get the hell out of the way. If you're going the same speed or less as the cars on the right, you aren't passing. If you want to trot out the law and say that you're just obeying the speed limit, you should know that in many states driving in the left lane is only permitted when passing. In other words, get the hell out of the way.
Yeah, because some line manager who is trying to do something simple like track productivity and attendance of his employees has no business spending a few hours creating a Access database to meet his needs.
The proper way of doing things is to submit a technology project request to IT, wait a couple weeks for a project manager to be assigned, spend a few days putting together a requirements document, wait a few weeks for the requirements document to be reviewed, spend a couple more days rewriting the requirments document, wait a few more weeks for the project to be prioritized, and then wait another 3-5 years for the project to actually be completed, if it doesn't get delayed even further because of projects with a higher priority.
As a database guy in a large corporation, I think it's great that employees can create small things in Access. It frees up my time for projects that are important and challenging. And when the Access databases actually become buisness critical, I migrate them to SQL Server or Oracle. Since you'd clearly prefer to be the bottleneck preventing people from helping themselves, I'm glad you don't set policy where I work.
No. If Google wanted to start screwing people, they could demand that Bell South start paying Google money so that searches from Bell South customers could be "prioritized". If you don't think Google has enough clout, what about ESPN or NASCAR? They could block Bell South customers unless Bell South paid for the right to carry their content.
We have a standard way of doing applications over the internet, it's designed for doing applications over the internet, and the only reason it isn't optimized for the task is because some people just can't seem to move past 1993 and constantly whine that that isn't what the web was designed for.
Well, tough shit. That's what the web is being used for. Do you really think that the web would have been successful without ecommerce? Ecommerce is a web application. Any site that takes input from the user and does something with it is a web application. Slashdot is a web application.
The whole point of DOM, DHTML, scrpting, XML, CSS, and practically every web standard for the last 7 years has been to facillitate using the web for applications. Just about everyone is on board with this, but there's always a few that want to retard progress.
Web applications do exist, they aren't going away, and any attempt to change this fact is just pissing into the wind. Stop being part of precipitate and start being part of the solution.
Criminal charges don't make sense, but how about revoking their corporate charter? Sure, the investors are out of luck and lots of employees would be out of jobs, but that gives everyone the incentive to make sure that illegal activity doesn't take place.
The PS2 hard drive didn't sell for shit because Sony crippled it intentionally. The only reason they released it in the states is because FFXI can't be played without it. Many other games that had hard drive support in Japan were forced to disable that support when they were localized for NA.
You want the "best" HA solution but not something too expensive. How about you give us something more to go on, like how much are you willing to spend, how much downtime you can tolerate in the event of failure, are there space/power constraints, etc. Then people can give you a real recomendation instead of the standard MySQL sucks/is great.
A threat of harm can't be financial? Your boss says do it or you'll be fired, most people will go ahead and do it. Then the ones with principles might starting looking for another job. Standing up for the right thing isn't always the best move when you have a family to support.
What does Visual Source Safe have to do with Office documents? Are you trying to put Power Point presentations under source control? A document repository with versioning capabilities might be more appropriate. Sharepoint Services and Domino.doc both come to mind.
If you want to manipulate Office documents in some arbitrary program, you can always save them as XML. Anything that can read/write text files gives you the ability to manipulate the document.
Or you can always write some in-house code. The native format for Office is BIFF. It's not terribly hard to figure out.
since in the nonsubjective ways reading electronically is pretty much universally better
And I thought reading for pleasure was pretty much entirely a subjective experience.
When I want information, I go to the electronic version. When I want to relax, the dead tree version is the only way to read. The subjective reasons you dismiss so quickly all center around engaging additional senses. If you don't understand that touch and smell can enhance pleasure, all I can suggest is to find a girlfriend or boyfriend and see if that clears up a few things for you.
Book publishers are still in trouble, but not because electronic books are better than paper. They're in trouble because one-off printing on demand at an affordable price isn't very far in the future. The author could sell their book in electronic form on the net and the customer could send it over to the local print shop for printing and binding and the total could be substantially less than the major publishers charge for a hardcover now. That's going to shrink their margins substantially. They'll have to become leaner and more agressive about attracting upcoming young authors.
It has a complete, robust security model. You also have to do a lot of reconfiguring to actually use said security model. Now, I could accept that on XP Home. XP Pro is the same, though. Create an account. Make that account a power user instead of a regular user or administrator. Log into that account and watch everything fall apart. That's why many businesses have chosen to stay on 2000.
No. The real problem with big big big public companies is that senior executives give themselves huge stock options, sacrifice long term profitability for short term gain, exercise their options, diversify their portfolio, and take a position elsewhere before they have to answer for their actions.
LOL. Ok, I'll buy that. Still the new controllers are easier than those damn Colecovision monstrosities. My experience has been the opposite. Most of my peer group plays games on the new consoles because it's something they can do with their kids. Many of the college students I meet don't have a clue how to play video games.
Why do you believe a successful console requires Japanese games? The vast majority of consoles are not sold in Japan. Is it that you feel that the Japanese are the only developers creating good games? Or maybe it's that western developers are still more focused on the PC than consoles? In either case, it would probably be a lot easier for Microsoft to encourage western developers to produce more/better console games than to overcome the built in prejudice the Japanese have towards foreign companies.
Slashdot Mirror
I don't care how good a computer scientist you are, the code you write in the first year or two you use a particular language is crap.
Sure, you understand the fundamentals of computer languages like flow control and logical operators, but you don't have a practical understanding of how to best use features of a particular language. A couple days with a book lets you write programs that you look back on a few years later and wonder "What the hell was I thinking when I wrote that?"
I have yet to meet a computer scientist right out of college that truly understands the tools used in information systems. They may or may not know how to write programs in languages used in the business world. They might have a limited grasp of relational databases but they've probably never encountered multi-dimensional databases. They certainly never bothered with anything as antiquated as mainframes.
It takes years of experience to become a "real" computer scientist instead of just another bozo with a sheepskin.
A properly trained professional with advanced experience will write a better product. I agree completely.
The problem is that not only do you confuse education with traiing and experience, you misunderstand the nature of computer science. Computer scientists learn about things like operating system design, language theory, hardware architecture and artificial intelligence. None of those things are applicable to writing business applications.
For business applications you're better off hiring someone with an information systems background. They've learned about how computers are used in the real world and have probably done an internship where they've had to work with business applications.
You missed the point. Some of the time, you don't need super speed as much as you need inexpensive to maintain.
Oh, and you should be using C instead of C++. Sure the overhead your talking about isn't much but why waste the hardware on overhead when you can take full advantage of it.
You don't need to reimplement COM and OLE just to parse the file. At most you might have to reimplement part of the Storage API to get to the streams. And you'd only have to do that if you were A) targeting a non-Windows platform and B) didn't want to grab some existing code from an open-source project like Calc.
You might need COM and OLE to display the file as it would appear in Office, but not to parse it.
Smaller tags = smaller file = faster reads. Smaller tags = fewer operations required to distinguish tags = faster parsing. The speed of string comparisons is always determined by the length of the strings. Checking if a tag matches 'somereallylongdescriptivetagname' will always take longer than comparing against 's'. There are some optimizations you can do based on the hardware. You can compare 4 case sensitive letters at a time on 32-bit hardware or 8 case sensitive letters at a time on 64-bit hardware. For case insensitive comparisons like xml, you're pretty much stuck with letter at a time.
As for why CSV isn't satisfactory for spreadsheets, make friends with an accountant or financial analyst and get them to show you how they use formulas, what-if scenarios, pivot-tables, and all of the other features of a modern spreadsheet program. CSV with a statistical analysis program is great for scientific data, but it doesn't cut it for financials.
It's nice for small shops. In a large corporation with 250,000+ employees, Sharepoint doesn't work so well.
I won't comment on the "collaboration" features, but a big part of Sharepoint and Domino are document repositories. Both of them have some major drawbacks in that area.
The big end user problem in Domino is the limited hierarchy. You have rooms, cabinets, folders, and files. People find it very limiting and confusing to be forced into that model. Some other issues involve poor performance and difficulty of performing backups because of certain design decisions that haven't been changed in the last 10 years. Everyone I've spoken to in my company hates using Domino.
Sharepoint offers unlimited hierarchy. The big problem in Sharepoint is security. You can set security on a respository but not on folders or documents. As far as I can determine, Windows authentication is required. This can be a real problem in a large corporation where various parts of the business have their own domains or active directory trees that aren't configured to trust the other domains or directories. Also, documents are differentiated and versioned entirely based on filename.
Sure, there are some things about document management that can be hard. I know from experience. I've written a document management system. That's why it amazes me that IBM and Microsoft haven't been able to put forth better offerings.
Interesting that your biggest problem is IE. My biggest problem is the lack of modal windows in Firefox and earlier versions of Safari. IE is the corporate standard, but employees working from home over VPN sometimes use other browsers. Many intranet applications require the use of modal dialogs and half-ass attempts to simulate modality do not make the cut.
IT wants to switch to Firefox but can't because Firefox doesn't provide a way to do a true modal dialog. There seems to be some kind of religious belief that modal windows are Evil but modals from alert(), confirm(), and prompt() are ok. I don't understand the theological difference between the two, but what I do understand is that the method Firefox's evangelism team advocates instead of showModalDialog does not offer the same functionality and tends to increase development and QA time.
We give project stakeholders the choice up front. We present their options in an unbiased fashion and let them decide. On every single project, the choice has been to use showModalDialog because it enables a consistent user interface without trying to anticipate and block all the different ways to navigate in the parent window.
Just about everything else can be made to work in every browser without much work. If the Firefox developers would follow in Safari's footsteps and implement showModalDialog, I could write applications that work in every major browser without compromising functionality or appearance. Then I might be complaining about the lack of CSS attributes like min-width in IE6. Unfortunately I think IE is going to catch up on the rendering side before Firefox implements anything because of business needs.
Clearly you don't work for a large organization. Applying a critical patch within week after it becomes available simply does not happen. Within a month, maybe. More likely 2-3 months. A regular patch takes 1-2 years. This is because the patches have to be tested against various configurations before being dissected and repackaged. Then the repakaged patch has to be tested against various configurations. The repackaging and testing occurs several times because someone always decides that they know better than the vendor which bits are actually required and leaves something out which later on turns out to be important.
We've been hit by a virus because some senior executive in one of the Manhattan offices decided to ignore policy and not put his internet web servers in the DMZ. Unfortunately, he's too powerful to get more than a slap on wrist. After the first attack I configured my servers to not accept any traffic from that block of addresses. I didn't patch my servers for about 3 months because no one is allowed to install anything that hasn't been repackaged and added to the standard operating environment. Senior management can ignore their own policies, but heaven help anyone else.
Sound stupid? It is. But it's par for the course in any really large organization.
Most of my consumer electronics are Sony. Receiver, DVD, TV, Playstation, PS2, etc. I have a rack full of PS and PS2 games and accessories. When I bought my PS2 on launch day, I thought $300 was a bit pricey. I could justify it to myself because stand alone DVD players were similar in price. I already had a DVD player but it would be nice to have one for the second TV as well. DVD offered clearly better picture quality than VHS on the TVs I already owned.
I have zero interest in a Blu-Ray or HD-DVD stand-alone player. I don't have an HDTV because HD programs only started being shown in my area two years ago. Even now, the selection of HD programming is very slim. My TVs probably have at least 5 years left on them before I start thinking about replacing them. Now, I'd be interested in a Blu-Ray or HD-DVD drive for my computer, but that's a different story.
So, I don't have HDTV. My existing DVD player outputs higher resolution than my TV can display. I have no desire for a Blu-Ray player. Why am I going to spend $500-$750 on a PS3? I'm not. I don't care how many killer games get released as PS3 exclusives. Until the price is $300 or less, I won't even consider it. Even then I still might not since I'm not happy about DRM.
If the PS3 costs substantially more than the XBox, Sony loses this generation of the consoles. It gets too expensive and Nintendo might even make it back into 2nd place.
Ok. PASS the car on the right and then get the hell out of the way. If you're going the same speed or less as the cars on the right, you aren't passing. If you want to trot out the law and say that you're just obeying the speed limit, you should know that in many states driving in the left lane is only permitted when passing. In other words, get the hell out of the way.
Yeah, because some line manager who is trying to do something simple like track productivity and attendance of his employees has no business spending a few hours creating a Access database to meet his needs.
The proper way of doing things is to submit a technology project request to IT, wait a couple weeks for a project manager to be assigned, spend a few days putting together a requirements document, wait a few weeks for the requirements document to be reviewed, spend a couple more days rewriting the requirments document, wait a few more weeks for the project to be prioritized, and then wait another 3-5 years for the project to actually be completed, if it doesn't get delayed even further because of projects with a higher priority.
As a database guy in a large corporation, I think it's great that employees can create small things in Access. It frees up my time for projects that are important and challenging. And when the Access databases actually become buisness critical, I migrate them to SQL Server or Oracle. Since you'd clearly prefer to be the bottleneck preventing people from helping themselves, I'm glad you don't set policy where I work.
No. If Google wanted to start screwing people, they could demand that Bell South start paying Google money so that searches from Bell South customers could be "prioritized". If you don't think Google has enough clout, what about ESPN or NASCAR? They could block Bell South customers unless Bell South paid for the right to carry their content.
We have a standard way of doing applications over the internet, it's designed for doing applications over the internet, and the only reason it isn't optimized for the task is because some people just can't seem to move past 1993 and constantly whine that that isn't what the web was designed for.
Well, tough shit. That's what the web is being used for. Do you really think that the web would have been successful without ecommerce? Ecommerce is a web application. Any site that takes input from the user and does something with it is a web application. Slashdot is a web application.
The whole point of DOM, DHTML, scrpting, XML, CSS, and practically every web standard for the last 7 years has been to facillitate using the web for applications. Just about everyone is on board with this, but there's always a few that want to retard progress.
Web applications do exist, they aren't going away, and any attempt to change this fact is just pissing into the wind. Stop being part of precipitate and start being part of the solution.
Criminal charges don't make sense, but how about revoking their corporate charter? Sure, the investors are out of luck and lots of employees would be out of jobs, but that gives everyone the incentive to make sure that illegal activity doesn't take place.
The PS2 hard drive didn't sell for shit because Sony crippled it intentionally. The only reason they released it in the states is because FFXI can't be played without it. Many other games that had hard drive support in Japan were forced to disable that support when they were localized for NA.
You want the "best" HA solution but not something too expensive. How about you give us something more to go on, like how much are you willing to spend, how much downtime you can tolerate in the event of failure, are there space/power constraints, etc. Then people can give you a real recomendation instead of the standard MySQL sucks/is great.
A threat of harm can't be financial? Your boss says do it or you'll be fired, most people will go ahead and do it. Then the ones with principles might starting looking for another job. Standing up for the right thing isn't always the best move when you have a family to support.
What does Visual Source Safe have to do with Office documents? Are you trying to put Power Point presentations under source control? A document repository with versioning capabilities might be more appropriate. Sharepoint Services and Domino.doc both come to mind.
If you want to manipulate Office documents in some arbitrary program, you can always save them as XML. Anything that can read/write text files gives you the ability to manipulate the document.
Or you can always write some in-house code. The native format for Office is BIFF. It's not terribly hard to figure out.
since in the nonsubjective ways reading electronically is pretty much universally better
And I thought reading for pleasure was pretty much entirely a subjective experience.
When I want information, I go to the electronic version. When I want to relax, the dead tree version is the only way to read. The subjective reasons you dismiss so quickly all center around engaging additional senses. If you don't understand that touch and smell can enhance pleasure, all I can suggest is to find a girlfriend or boyfriend and see if that clears up a few things for you.
Book publishers are still in trouble, but not because electronic books are better than paper. They're in trouble because one-off printing on demand at an affordable price isn't very far in the future. The author could sell their book in electronic form on the net and the customer could send it over to the local print shop for printing and binding and the total could be substantially less than the major publishers charge for a hardcover now. That's going to shrink their margins substantially. They'll have to become leaner and more agressive about attracting upcoming young authors.
It has a complete, robust security model. You also have to do a lot of reconfiguring to actually use said security model. Now, I could accept that on XP Home. XP Pro is the same, though. Create an account. Make that account a power user instead of a regular user or administrator. Log into that account and watch everything fall apart. That's why many businesses have chosen to stay on 2000.
No. The real problem with big big big public companies is that senior executives give themselves huge stock options, sacrifice long term profitability for short term gain, exercise their options, diversify their portfolio, and take a position elsewhere before they have to answer for their actions.
LOL. Ok, I'll buy that. Still the new controllers are easier than those damn Colecovision monstrosities. My experience has been the opposite. Most of my peer group plays games on the new consoles because it's something they can do with their kids. Many of the college students I meet don't have a clue how to play video games.
I'm guessing that your real question isn't how to use Oracle, but how to create a geodatabase. You might try "Designing Geodatabases: Case Studies in GIS Data Modeling"u seaction=display&websiteID=85&moduleID=0/
http://gis.esri.com/esripress/display/index.cfm?f
A google search on geodatabases turns up a number of similar resources.
Why do you believe a successful console requires Japanese games? The vast majority of consoles are not sold in Japan. Is it that you feel that the Japanese are the only developers creating good games? Or maybe it's that western developers are still more focused on the PC than consoles? In either case, it would probably be a lot easier for Microsoft to encourage western developers to produce more/better console games than to overcome the built in prejudice the Japanese have towards foreign companies.