The notion of Microsoft encouraging standards makes sense only in a universe in which CP/M did not exist before Microsoft DOS.
Before DOS, I could buy a computer from Xerox, Morrow, Kaypro, Osborne, and a dozen or so other companies. They all ran the same OS, they all ran on compatible CPUs, and guess what? There was open-source software for them.
Market forces, not Microsoft, spread the gospel of standardization. IBM, not Microsoft, legitimized microcomputing in the business community.
If anything, Microsoft's business practices have prevented standardization. I can swap hard drives and video cards, but I can't swap operating systems out from under my applications. (Wine isn't quite good enough for that yet.)
Slackware was NOT the first commercial distribution of Linux. SLS was there long before Slackware, back in the kernel 0.96 days, maybe even earlier. Like Slackware, it came as a big pile of.tgz files with 8+3 filenames so that any idjit could download it onto MS-DOS floppies (about 20, as I recall).
SLS cratered before the kernel hit 1.0. But it proved the concept.
After a week with OS X.1, I'm ready to reformat and start over. But there are no ISO images in the Yellow Dog archives. ??
I continue to be puzzled...
on
Mozilla 0.9.5
·
· Score: 1
I continue to be puzzled by posts proclaiming that Mozilla is faster than browser X. I've tested release after release against an array of browsers on multiple platforms -- even bizarre browsers such as one based on interpreted TCL -- and it just ain't so.
It's heartening that the current release actually runs, most of the time, and renders pages about as well as Internet Explorer.
But it's like wandering through the world socked on depressants. Even little things, like typing into the location bar, or clicking on a dropdown, are noticably slow.
Under Linux, the news and mail clients are useless.
Galeon may be the answer. I don't know. It's such a bugwad of undisclosed dependencies that I've never been able to get it running.
Whoa! That's not at all what the cited document says. It is clear:
"What is a work made for hire? Although the general rule is that the person who creates the work is its author, there is an exception to that principle; the exception is a work made for hire, which is a work prepared by an employee within the scope of his or her employment; or a work specially ordered or commissioned in certain specified circumstances. When a work qualifies as a work made for hire, the employer or commissioning party is considered to be the author. See Circular 9."
O Fortune,
like the moon
you are changeable,
ever waxing
and waning;
hateful life
first oppresses
and then soothes
as fancy takes it;
first posting
and taking away
just an April fool's joke.
It doesn't make any difference whether the product is open source or closed source; you should always search for trademark issues before naming a product. Two steps:
1. Use the US Patent and Trademark search engine at http://www.uspto.gov/. There may be similar resources in other countries.
2. Do a Web search. This is not a substitute for the search of registered trademarks, but it may help you avoid other bonehead maneuvers.
Regardless of whether you think patent and trademark laws are being abused by corporate interests, et cetera, you should perform these steps out of respect for your potential users. Well-chosen names will help people find your (free or not) product.
Re:Finally no more battles with these php freaks
on
Mason 1.0 Released
·
· Score: 2
You're wrong. It's not a strike against PHP. Here's why.
All templating systems require code to be embedded in HTML. It may be as simple as "put this variable here" -- but it's still code. A robust templating system that supports dynamic page generation, personalization, and database interaction will inevitably require common programming-language constructs, such as testing, looping and branching, at the presentation level.
While it is sound practice to separate business logic from presentation logic, and presentation logic from look and feel, it is a practice that should be followed insofar as it is practical and beneficial. And no further.
It's a lot like Einstein's position on simplicity: a theory should be as simple as possible, but no simpler. Do not follow ideas over cliffs.
By providing a common language at all layers of an application, PHP allows the application designer the freedom to do the right thing, or the wrong thing. I'm fine with that. The smart designer will make a wise choice, and the dumb designer is probably using a different platform anyway.
No, you don't have to code separately for each database -- you can use the PEAR portability objects that PHP includes (and ZDnet completely missed!), or you can use the popular third-party library PHPlib.
But you probably WANT to code separately. If you shoot for the least common denominator, you get the smallest feature set. ZDnet includes a sidebar about performance tuning; every DB-related suggestion breaks the entire portability model. Databases are all different!
I found the article shallow and curiously lacking in any real performance metrics that might help me understand the differences between the languages.
I only have up-close-and-personal familiarity with PHP. I found ZDNet's scoring odd:
"C" for developer time. ZDnet asserts that Cold Fusion, a proprietary language, is easier to learn. Well, that might be if you've never written any code in C or shell scripts. If you have, PHP is as comfortable as an old shoe. There's an awful lot of personal preference tied up in a score like this.
"D" for tools. Gee, no draw-and-drag programming environment. I don't get it. Maybe this is important to somebody. Have we spawned a generation of programmers who can't type?
"C" for installation. Maybe if you're installing it under Windows. Many Linux distributions come with PHP already installed. I did./configure, make, make install and it took about five minutes.
"C" for back-end connectivity. This is a tough one because there is a conflict between portability, on the one hand, and performance/features on the other. If you want a portable SQL API, you're going to be stuck with a least-common-denominator API. If you want to code for performance -- putting stored procedures in the DB, for example -- they're not going to run on some other RDBMS. You've got to make a choice, and PHP lets you do that -- there are abstract database API's, but the fact that few people actually use them is telling.
"A" for performance. Yeah, that would be important if you care about your users.
> The format Microsoft uses is the Open eBook standard set by the publishing industry.
The format that is documented on the openebook Web site is a text format. The Microsoft Reader files are not text. Where is there public documentation that allows me to build a reader for this "open" format?
Microsoft has cooked up a proprietary file format and a proprietary Windows-only ebook-reader that, if successful, would create yet another barrier to the open sharing of information.
See http://etext.lib.virginia.edu... where well over a thousand works in the public domain have been converted to Microsoft's proprietary format.
Among the "advantages" of Microsoft's format is encryption -- an anti-"piracy" measure. I don't know much about it, since Microsoft's Web page on that subject is all buggered up and won't display properly with Netscape.
There is an Open Ebook Forum that has documented a text format, but it has not documented anything about encryption, rights management, etc.
It is not merely the prospect of a single company controlling the "standard" that I find chilling; it is the opportunity for the standard to be abused in the ways the DVD cartel has abused its power. By companies, or by governments.
Imagine geocoding that keeps out bad thoughts from other countries.
Imagine some 14-year-old programmer getting arrested for making it possible to read.
I think there's something to that. WAP is a false Internet, a closed network masquerading as "wireless Web." It is a marketing ploy to differentiate a product that is fundamentally a commodity. The value isn't there.
There is real value in the vision held by many of the cellular providers, but we are a long way from that vision. The vision requires:
Better user interfaces. Doh! We will not triple-tap our way to http://wap.somehorrendouslylong.url.com.
Displays big enough to read and display useful graphics (not video, just maps and such).
GPS integration.
The idea is that when my Bridgestone/Firestone Tires of Death shred themselves, I can whip out my phone and get directions to the nearest service station, then locate the closest Starbuck's for a cup of coffee while they're replaced.
But we're a long way from that.
I've been watching with some amusement as American companies scramble to come up with WAP strategies. In general, these strategies have two key components:
Exclusivity, reflecting the "false Internet" nature of the so-called Wireless Web. If you're on the screen, you're on. If you're off the screen, you're dead. The telephone company owns the screen. Pay up, information providers, or miss out on the Next Big Thing. But nobody really knows whether this next thing is big. Where's the money for the information provider?
Weak content planning, generally spun out of the "we have all this great content, let's give it to them" school of thought rather than the "let's understand what a wireless consumer might need" school of thought. Some newspaper people, for instance, are talking about how great it would be to put classified ads on WAP. I don't think I am going to sit in the park and read the classifieds on my phone.
From the Financial Times: Dutch papers fail in internet copyright case: Leading Dutch newspapers yesterday failed to prevent an online
news service from providing direct links to articles on newspaper websites, in a legal ruling that helps define the limits of internet copyright.
It sounds like you're looking for something like Urbanite, which is a commercial turnkey suite of tools for building a Geocities-like system.
Most of the large players have either built their own tools or contracted with integrators to have tools built for them.
Platforms such as Midgard (based on PHP) and Zope (based on Python) make it radically easier to develop such tools.
There also are any number of open-source Slashdot-like environments such as Squishdot; some browsing around in Freshmeat.net will churn them up quickly.
Not that I want to interrupt the bashAOL festival, but when I logged onto AOL today (I use it when traveling) the following tease was featured on the first screen:
Upgrading to Windows2000? Click here for important information you should know first.
It links to a screen that warns "the AOL service is not supported on this or other WindowsNT platforms at this time."
Users can ask to be emailed when an NT-compatible version is available.
Slashdot Mirror
All @Home services are delivered through cable "partners," including Cox, Comcast, AT&T, et cetera. @Home has no independent customers.
Yes, Cox@Home is affected.
(Disclosure: I used to work for Cox Interactive, and I own a small amount of now-worthless @Home stock.)
Maybe this will attenuate the NIMBDA attacks from cable-connected Windows PCs.
It's not about technology.
It's not about fairness.
It's always about power. Always.
The notion of Microsoft encouraging standards makes sense only in a universe in which CP/M did not exist before Microsoft DOS.
Before DOS, I could buy a computer from Xerox, Morrow, Kaypro, Osborne, and a dozen or so other companies. They all ran the same OS, they all ran on compatible CPUs, and guess what? There was open-source software for them.
Market forces, not Microsoft, spread the gospel of standardization. IBM, not Microsoft, legitimized microcomputing in the business community.
If anything, Microsoft's business practices have prevented standardization. I can swap hard drives and video cards, but I can't swap operating systems out from under my applications. (Wine isn't quite good enough for that yet.)
The new antitrust policy is inspired by a successful program from the Fish and Wildlife Service. It's called "catch and release."
Slackware was NOT the first commercial distribution of Linux. SLS was there long before Slackware, back in the kernel 0.96 days, maybe even earlier. Like Slackware, it came as a big pile of .tgz files with 8+3 filenames so that any idjit could download it onto MS-DOS floppies (about 20, as I recall).
SLS cratered before the kernel hit 1.0. But it proved the concept.
After a week with OS X.1, I'm ready to reformat and start over. But there are no ISO images in the Yellow Dog archives. ??
I continue to be puzzled by posts proclaiming that Mozilla is faster than browser X. I've tested release after release against an array of browsers on multiple platforms -- even bizarre browsers such as one based on interpreted TCL -- and it just ain't so.
It's heartening that the current release actually runs, most of the time, and renders pages about as well as Internet Explorer.
But it's like wandering through the world socked on depressants. Even little things, like typing into the location bar, or clicking on a dropdown, are noticably slow.
Under Linux, the news and mail clients are useless.
Galeon may be the answer. I don't know. It's such a bugwad of undisclosed dependencies that I've never been able to get it running.
Whoa! That's not at all what the cited document says. It is clear:
"What is a work made for hire? Although the general rule is that the person who creates the work is its author, there is an exception to that principle; the exception is a work made for hire, which is a work prepared by an employee within the scope of his or her employment; or a work specially ordered or commissioned in certain specified circumstances. When a work qualifies as a work made for hire, the employer or commissioning party is considered to be the author. See Circular 9."
O Fortune,
like the moon
you are changeable,
ever waxing
and waning;
hateful life
first oppresses
and then soothes
as fancy takes it;
first posting
and taking away
just an April fool's joke.
1. Use the US Patent and Trademark search engine at http://www.uspto.gov/. There may be similar resources in other countries.
2. Do a Web search. This is not a substitute for the search of registered trademarks, but it may help you avoid other bonehead maneuvers.
Regardless of whether you think patent and trademark laws are being abused by corporate interests, et cetera, you should perform these steps out of respect for your potential users. Well-chosen names will help people find your (free or not) product.
You're wrong. It's not a strike against PHP. Here's why.
All templating systems require code to be embedded in HTML. It may be as simple as "put this variable here" -- but it's still code. A robust templating system that supports dynamic page generation, personalization, and database interaction will inevitably require common programming-language constructs, such as testing, looping and branching, at the presentation level.
While it is sound practice to separate business logic from presentation logic, and presentation logic from look and feel, it is a practice that should be followed insofar as it is practical and beneficial. And no further.
It's a lot like Einstein's position on simplicity: a theory should be as simple as possible, but no simpler. Do not follow ideas over cliffs.
By providing a common language at all layers of an application, PHP allows the application designer the freedom to do the right thing, or the wrong thing. I'm fine with that. The smart designer will make a wise choice, and the dumb designer is probably using a different platform anyway.
I want to know if it rubs off on my hands.
But you probably WANT to code separately. If you shoot for the least common denominator, you get the smallest feature set. ZDnet includes a sidebar about performance tuning; every DB-related suggestion breaks the entire portability model. Databases are all different!
I only have up-close-and-personal familiarity with PHP. I found ZDNet's scoring odd:
"C" for developer time. ZDnet asserts that Cold Fusion, a proprietary language, is easier to learn. Well, that might be if you've never written any code in C or shell scripts. If you have, PHP is as comfortable as an old shoe. There's an awful lot of personal preference tied up in a score like this.
"D" for tools. Gee, no draw-and-drag programming environment. I don't get it. Maybe this is important to somebody. Have we spawned a generation of programmers who can't type?
"C" for installation. Maybe if you're installing it under Windows. Many Linux distributions come with PHP already installed. I did ./configure, make, make install and it took about five minutes.
"C" for back-end connectivity. This is a tough one because there is a conflict between portability, on the one hand, and performance/features on the other. If you want a portable SQL API, you're going to be stuck with a least-common-denominator API. If you want to code for performance -- putting stored procedures in the DB, for example -- they're not going to run on some other RDBMS. You've got to make a choice, and PHP lets you do that -- there are abstract database API's, but the fact that few people actually use them is telling.
"A" for performance. Yeah, that would be important if you care about your users.
I used to think it meant "My Operating System Is A Dead Parrot."
I wonder how many of Slashdot's born-yesterday users even know what you're talking about. Robert T. Morris? Internet worm? :-)
It gets() depressing.
The format that is documented on the openebook Web site is a text format. The Microsoft Reader files are not text. Where is there public documentation that allows me to build a reader for this "open" format?
I didn't say they weren't.
Once upon a time all video was available in Betamax format.
Microsoft has cooked up a proprietary file format and a proprietary Windows-only ebook-reader that, if successful, would create yet another barrier to the open sharing of information.
See http://etext.lib.virginia.edu ... where well over a thousand works in the public domain have been converted to Microsoft's proprietary format.
Among the "advantages" of Microsoft's format is encryption -- an anti-"piracy" measure. I don't know much about it, since Microsoft's Web page on that subject is all buggered up and won't display properly with Netscape.
There is an Open Ebook Forum that has documented a text format, but it has not documented anything about encryption, rights management, etc.
It is not merely the prospect of a single company controlling the "standard" that I find chilling; it is the opportunity for the standard to be abused in the ways the DVD cartel has abused its power. By companies, or by governments.
Imagine geocoding that keeps out bad thoughts from other countries.
Imagine some 14-year-old programmer getting arrested for making it possible to read.
See Jakob Nielsen's July 9 column, WAP Backlash.
There is real value in the vision held by many of the cellular providers, but we are a long way from that vision. The vision requires:
Better user interfaces. Doh! We will not triple-tap our way to http://wap.somehorrendouslylong.url.com.
Displays big enough to read and display useful graphics (not video, just maps and such).
GPS integration.
The idea is that when my Bridgestone/Firestone Tires of Death shred themselves, I can whip out my phone and get directions to the nearest service station, then locate the closest Starbuck's for a cup of coffee while they're replaced.
But we're a long way from that.
I've been watching with some amusement as American companies scramble to come up with WAP strategies. In general, these strategies have two key components:
Exclusivity, reflecting the "false Internet" nature of the so-called Wireless Web. If you're on the screen, you're on. If you're off the screen, you're dead. The telephone company owns the screen. Pay up, information providers, or miss out on the Next Big Thing. But nobody really knows whether this next thing is big. Where's the money for the information provider?
Weak content planning, generally spun out of the "we have all this great content, let's give it to them" school of thought rather than the "let's understand what a wireless consumer might need" school of thought. Some newspaper people, for instance, are talking about how great it would be to put classified ads on WAP. I don't think I am going to sit in the park and read the classifieds on my phone.
From the Financial Times: Dutch papers fail in internet copyright case : Leading Dutch newspapers yesterday failed to prevent an online news service from providing direct links to articles on newspaper websites, in a legal ruling that helps define the limits of internet copyright.
Most of the large players have either built their own tools or contracted with integrators to have tools built for them.
Platforms such as Midgard (based on PHP) and Zope (based on Python) make it radically easier to develop such tools.
There also are any number of open-source Slashdot-like environments such as Squishdot; some browsing around in Freshmeat.net will churn them up quickly.
Not that I want to interrupt the bashAOL festival, but when I logged onto AOL today (I use it when traveling) the following tease was featured on the first screen:
Upgrading to Windows2000?
Click here for important information
you should know first.
It links to a screen that warns "the AOL service is not supported on this or other WindowsNT platforms at this time."
Users can ask to be emailed when an NT-compatible version is available.
* Slide show: Battle in Seattle
* Slide show: Mobs and mayhem
* Live Seattle traffic cams
* Pine Street
* Live traffic conditions
* Road closures