my own "yet another imageboard-script" (116)

1 Name: #!/usr/bin/anonymous : 2007-02-26 01:26 ID:Hw07jnmQ This thread was merged from the former /code/ board. You can view the archive here.

So - I promised a friend to set up a imageboard for him, but I quickly realized that the only good (open source) imageboard out there is written in Perl.

I didn't take long before I got hooked on the idea of writing my own board.
I have to admit I'm a ruby fanboy (no flameing please :) ), but that doesn't change the fact that 90% of webhosts out there only supports PHP, so I decided to write it in PHP5/MySQL5.

Yesterday I had my first, sort of, working release which I really would like some feedback on.

It's not, in any way, intended to be used in real life yet, but I some of you would help me download it, test it, report problems, give feedback, suggestions etc. etc. etc. It would be really cool :)

thanks in advance

oh yeah - and the URL to the project page is: http://code.google.com/p/4chandk/

67 Name: #!/usr/bin/anonymous : 2008-09-07 04:21 ID:gHwzGZFx

>Here you go, no concrete statements; hinging on your private and secret definition of "widely available".

Because I new someone would go and point out some little known browser that didn't serve as a practical browser for the mass market.

>Arena had CSS support and editing.

See. Arena was a test platform. Not a mass market web browser.

>Without saying why this is good.

I said it was innovative, not that it had to be good.

>Wrong. Netscape's RDF channels (on the Netscape Home Page) predates IE4. IE4's channels were released to combat it.

IE6 was 1996. Netscape RDF was 1999.

>Wrong. Netscape 3 could do all of those things just as accidentally as IE5 could. Using liveconnect instead of ActiveX, a great many things were possible, and yet still it took someone else to figure out it was valuable.

I am not to sure. Using the Java plug-in with JavaScript isn't really the same as having a purpose built object available to all versions of the browser for the task. ActiveX was used for browser plug-ins, but that was not the sole implementation of the technology. XMLHTTPRequest isn't a plug-in.

IE5's implementation was no accident. MS used it to make a web interface for Outlook. GMail was not the first AJAX web based email.

>Who said they didn't?

The comment I pointed out earlier that was the whole point of that post.

>Why don't you challenge my argument?

Come up with a more practical use of forking. Forking the Linux kernel doesn't cut it. And make sure that your example makes it clear exactly why the features could never be added to the original project.

68 Name: dmpk2k!hinhT6kz2E : 2008-09-07 04:28 ID:Heaven

> Apple takes four years between releases

There's been five releases of OSX since 2001.

Also, I haven't really been reading earlier posts, but can you please remove the dick-sucking rhetoric from your posts in the future? It's possible to argue without being an ass, Internet or not.

69 Name: #!/usr/bin/anonymous : 2008-09-07 08:29 ID:Heaven

> Come up with a more practical use of forking.

forking works great to split my program into multiple processes.

70 Name: #!/usr/bin/anonymous : 2008-09-08 02:41 ID:qPrStbkM

> I am pretty sure that the Windows kernel will let you view its page faults.

"view its page faults"?

What are you doing on /code/ if you don't write software?

What was needed was a way to detect if a memory access would generate a page fault. That turns out to be very hard to do in user-space because the information is obsolete as soon as it's made available to user-space (say via an interface like mincore()).

Deep changes to Linux's vmm made this possible.

> The idea of a fork is to take one app and make it in to 2 completely separate apps.

Nope. That's one idea of a fork. If that's your only idea, then you're wrong.

You can fork to separate to be sure, or you can fork with the intention of merging later.

Some forks have no intention of living past the proof-of-concept stage, and some simply branch off to provide bugfixes to version-locked sites.

These are all forks. None of them necessarily lead to resource-contention because not all resources are equally usable. The resources that can be utilized to advance virtualization or garbage collection do not work in a vacuum. They may require radical changes to a system, and they might not work.

There may be political reasons why two people cannot work together- there may be social, or legal reasons.

So perhaps your whole problem is that you have no idea how forks work in reality. Forking for vanity- the kind of forking you're talking about, is a very rare, and very insignificant kind of forking. It causes very little harm (besides bruising egos) because the people doing it aren't necessarily capable of contributing together anyway.

> If you wanna say NeXT was started in 1985 and GNU started in 1984 then its not really that much different.

Ah, but there was a usable Linux desktop long before OSX came out.

Ergo: you are wrong, yet again.

> Linus started his work in 1991 though. OS X came from next, but its actual development as the OS we know today isn't the same thing they started in 1985.

Then do you want to measure it by the time between releases?

Or are you going to continue to make baseless arguments that you don't have to defend?

71 Name: #!/usr/bin/anonymous : 2008-09-08 02:42 ID:qPrStbkM

> > Their [MS's] web browser saw zero innovation in the near-decade they were champ.

I still stand by this. You didn't bring up a single thing "innovated" during this decade.

Not one.

IE3? Really?

> But you need me to set you straight.

I do. Now please do it. I've been waiting this entire time.

You have failed so miserably up until this point. You've made no definitive remark besides forking is always bad, and haven't backed that up with anything substantial.

Try something resembling logic; it begins with a statement like this:

Mac OSX, an example of why forks are always bad, begin development in my cockeyed mind on date X, and ended on date Y, thus taking less time than the first usable Linux desktop started on date Q, ending on date P.

Here's another one: Come up with a single Microsoft innovation that isn't in response to market forces.

You already stated that there shouldn't be any, which demonstrates my point: forks make it easier for people to experiment. Thanks for not only being wrong, but demonstrating I'm right.

Meanwhile, my point needs little demonstration; it is simple: forking is sometimes good because good things have happened as a result of forks.

72 Name: #!/usr/bin/anonymous : 2008-09-08 02:45 ID:Heaven

> Also, I haven't really been reading earlier posts, but can you please remove the dick-sucking rhetoric from your posts in the future? It's possible to argue without being an ass, Internet or not.

For what it's worth, I think I started off extremely sociable in >>52 , and not once in this thread mentioned "dick sucking".

73 Name: #!/usr/bin/anonymous : 2008-09-08 03:44 ID:gHwzGZFx

>What are you doing on /code/ if you don't write software?

So the grad student definition of someone who writes code is a person who implements their own garbage collector with a page fault detection algorithm and then poorly explains this.

>What was needed was a way to detect if a memory access would generate a page fault.

Why? Are you marking things for GC when they are not present in memory? That can't be it, because it would make no sense to collect something just because its stored external to memory.

>That turns out to be very hard to do in user-space because the information is obsolete as soon as it's made available to user-space (say via an interface like mincore()).

I would imagine it would be. Apps usually don't care if their data is in memory or stored externally at any given time because they rely on the kernel to manage this.

Initially, I can't think of why one would write a garbage collector that uses a page fault algorithm to determine if information is available for collection based on if it is in memory or external to it.

>Nope. That's one idea of a fork. If that's your only idea, then you're wrong.

It's my idea of a good fork.

>or you can fork with the intention of merging later

Not as good, but not that bad either. I am not sure that would even be a fork. Forks create separate projects. In this scenario, we start with one project and maintain the intent to finish with one project. Sounds more like branching.

>Some forks have no intention of living past the proof-of-concept stage

Not really what I was talking about. But surely that would be a fork.

>and some simply branch off to provide bugfixes to version-locked sites.

Pretty sure that is considered a branch.

If we are gonna call branching a type of fork, then we can just exclude branching because that usually serves a logical purpose and the very idea is to more effectively manage resources.

>There may be political reasons why two people cannot work together- there may be social, or legal reasons.

To me, these all fall in to bad reasons to fork. Basically, boo hoo get over it. Except for legal, why would their be a reason to fork based on legalities of purely open source software?

>Ah, but there was a usable Linux desktop long before OSX came out.

I wasn't saying use-able. I was saying good. X predates NeXTs Display PostScript. But OSX doesn't even use that. The best Linux desktops currently available are Gnome (1997) and KDE (1996). Apple bought NeXT in 1997 and release OS X in 2001. OS X was already a better desktop OS after those 4 years of development than either KDE or Gnome was after the same time period. Even if you go back and count all the things that went before each development effort in to these desktops, OS X was better faster. Or if you expand it to the idea of entire distros, OS X still wins.

>Then do you want to measure it by the time between releases?

Each releases time and features are completley arbitrary. Not every incremental release introduces wide ranging features or enhancements. There is no standard shared between the 2. Comparison like that isn't relevant. A valid comparison would be the start of the development effort to what the teams concider a major release. Or even easier in this case is start to current.

>Or are you going to continue to make baseless arguments that you don't have to defend?

As I have further demonstrated, there is no comparable difference between a Linux desktop and OS X that goes against my point that Apple made a better desktop OS faster than Linux.

74 Name: #!/usr/bin/anonymous : 2008-09-08 03:59 ID:gHwzGZFx

>I still stand by this. You didn't bring up a single thing "innovated" during this decade.

You go ahead. I did make a whole list, but you can chose to ignore that. I even made the anecdotal case about how the web isn't the same as it was 10 years ago. But even that you can ignore.

And you really do a good job of "standing by this" by accurately refuting absolutely nothing. Just ignoring the truth that inconveniently doesn't fit this predetermined conclusion.

>Try something resembling logic; it begins with a statement like this:
>Mac OSX, an example of why forks are always bad...

Mac made a great OS in less time that any Linux fork has been able to. Linux has many Desktop forks. OSX has none. Seems like all of that separate efforts combined in to one, or maybe even 2 or 3 efforts, would have been more effective. The main goal is a Desktop Linux (read: the same), and the efforts are very similar. The differences in their execution could be worked out in their implementation instead of effort focused on packets of isolation.

>Here's another one: Come up with a single Microsoft innovation that isn't in response to market forces.

This is a new point you brought up. You first should prove that MS must produce innovation absent of market forces.

>You already stated that there shouldn't be any, which demonstrates my point: forks make it easier for people to experiment.

Expirimental, proof-of-concept "forks" are not what I was talking about. I even went on to demonstrate that in my first statement. Arguing semantics doesn't invalidate my point.

>Meanwhile, my point needs little demonstration; it is simple: forking is sometimes good because good things have happened as a result of forks.

Please do give a little demonstration in the context I was talking about. Something good to happen because a fork was absolutely necessary.

75 Name: nother anon : 2008-09-08 11:18 ID:Heaven

Guys, you are arguing with anonymous strangers on the Internet. There is absoluteny no reason to have anything invested in being right here.

And you're doing it RONG^W in a thread from early 2007 about writing your own board scripts. Please stop? You already know neither of you is going to come around. And nobody is impressed with you being "tough on wrong".

76 Name: #!/usr/bin/anonymous : 2008-09-08 11:30 ID:Heaven

>>75
You're unable to write coherent English sentences, so please, lurk more, post less.

77 Name: #!/usr/bin/anonymous : 2008-09-08 14:54 ID:GJSJH4YG

> Initially, I can't think of why one would write a garbage collector that uses a page fault algorithm to determine if information is available for collection based on if it is in memory or external to it.

Page faults are very slow. Most garbage collection algorithms work poorly on paged systems because of it. The best appear to be tricolor marking systems but only because they tend to localize the page faults to an area that an adaptive vmm can cope with.

This is far from ideal, and does more to perpetuate the myth that garbage collection is slow than anything else.

That said, your own inadequacies are noted.

> we can just exclude branching because that usually serves a logical purpose and the very idea is to more effectively manage resources.

I hopefully am parsing this correctly: You reject calling this kind of thing "forking" because the goals aren't for differentiation?

> Basically, boo hoo get over it.

Good luck with that. It's not very practical. I'd rather have two developers compete to make better products (that can copy code back and forth), than to have neither of them work very hard at all.

> Except for legal, why would their be a reason to fork based on legalities of purely open source software?

See Wine.

> I wasn't saying use-able. I was saying good. X predates NeXTs Display PostScript. But OSX doesn't even use that. The best Linux desktops currently available are Gnome (1997) and KDE (1996). Apple bought NeXT in 1997 and release OS X in 2001. OS X was already a better desktop OS after those 4 years of development than either KDE or Gnome was after the same time period. Even if you go back and count all the things that went before each development effort in to these desktops, OS X was better faster. Or if you expand it to the idea of entire distros, OS X still wins.

So you're counting 1997-2001 for OSX because that's when Apple bought NeXT? and you're counting KDE from 1996 to when exactly?

I don't think what you're saying makes any sense.

78 Name: #!/usr/bin/anonymous : 2008-09-08 14:59 ID:GJSJH4YG

> And you really do a good job of "standing by this" by accurately refuting absolutely nothing.

I refuted all of it. Every one of your examples either predates the time-frame in-question, or is the direct result of Firefox or other similar market forces.

None of it is innovation; none of it is demonstration that Microsoft (or anyone else) can be trusted with protecting us from the right to fork.

> The differences in their execution could be worked out in their implementation instead of effort focused on packets of isolation.

I disagree. GNOME2 started a lot slower than KDE and Macosx. Now it's a lot faster and lighter, with no architectural shifts. KDE and Macosx haven't gotten faster or smaller without paradigm shifts- KDE moving to plasma, and MacOSX moving away from Carbon and the corba-like interface in NeXT.

Nobody's exactly sure what the best architecture will be; Ten years ago, many people would've predicted the KDE-like approach, and twenty-years ago the SOM-based OS/2. Today, GNOME seems to be doing the right thing. Competition is what fuels these discoveries.

> > Here's another one: Come up with a single Microsoft innovation that isn't in response to market forces.
> This is a new point you brought up. You first should prove that MS must produce innovation absent of market forces.

You're contending Microsoft continues to innovate, so we don't need the ability to experiment on our own, aren't you?

> Expirimental, proof-of-concept "forks" are not what I was talking about. I even went on to demonstrate that in my first statement. Arguing semantics doesn't invalidate my point.

Where do you draw the line then? Is Konqueror still an experimental fork of GtkHTML? Since Webkit is merging back into KHTML, is it still a proof-of-concept?

Is the ac-branch just mere proof-of-concept for the Linux kernel?

You're the one that started this argument that all forks are bad. Do you really want to pick such a narrow definition of fork to satisfy that?

> Please do give a little demonstration in the context I was talking about. Something good to happen because a fork was absolutely necessary.

That's a logical fallicy. If I say webkit (for example), you simply say it wasn't absolutely necessary for Apple to fork, and there's no way to disprove it because it isn't falsable.

79 Name: #!/usr/bin/anonymous : 2008-09-09 04:16 ID:gHwzGZFx

>That said, your own inadequacies are noted.

You still havn't explained how your algorithm uses page faults and why it is so vital.

Your inability to explain this makes me wonder exactly how much work you did on this project.

Are you trying to avoid loading paged out virtual memory to physical memory when your CG is running?

>I hopefully am parsing this correctly: You reject calling this kind of thing "forking" because the goals aren't for differentiation?

Or you could say I am calling it branching because it is known as branching. Branching is a type of fork, but you could easily infer from my original statement that I was not talking about this type of forking. What I was talking about is simply called forking.

>I'd rather have two developers compete to make better products (that can copy code back and forth), than to have neither of them work very hard at all.

Ideally, they wouldn't be such pussies. I guess it worked for Edison and Telsa. Let's hope all programmers are pasionate geniouses.

>See Wine.

HAHAHA. You are going to have to be more specific. If you are talking about Wine forks I can think of:
Cedega, which mostly underscores my point. They took Wine, made a proprietery version of it and refused to contribute back. So Cedega made their own enhancements and those enhancements took duplicate effort to implement in Wine. Of course MIT was pissed about this and changed the license. The activley fixed an unintended negative consequence of forking. Cedega and Wine try to accomplish the same things, but Wine does it better and that is not because of Cedega.

Then there is CrossOver, another proprietery fork. But many of the active developers of Wine work for the makers of CrossOver. For all intents and purposes they function as a commercial branch of Wine. And the statement "the developers can't work togther for legal reasons" does not apply since they do work together.

Wine likes forking to proprietery stuff. Is the best concequence of forking that it can end forking?

>So you're counting 1997-2001 for OSX because that's when Apple bought NeXT?

It's a safe assumption they bought it to make OSX. They only stated as much back when they bought it that they were purchasing it to make their next OS for Macs.

>and you're counting KDE from 1996 to when exactly?

Go from 1996 - 2000 (same time period) if you like. Still not as good. Go up until now, still not as good.

80 Name: #!/usr/bin/anonymous : 2008-09-09 04:57 ID:gHwzGZFx

>I refuted all of it. Every one of your examples either predates the time-frame in-question, or is the direct result of FireFox or other similar market forces.

Seems like FireFox was the only market force in the last 10 years, and it only accounts for a few of those years.

Other, but not similar, market forces came directly from consumers. FireFox has only been doing anything since like 1995. Good things happened on the browser (the MS browser) and web during those other 7 years where MS dominated.

>Nobody's exactly sure what the best architecture will be; Ten years ago, many people would've predicted the KDE-like approach, and twenty-years ago the SOM-based OS/2. Today, GNOME seems to be doing the right thing.

And yet, they share so much in common. And you act like there was only 1 way to go, one approach if forking wasn't the equation. Either the KDE way or the Gnome way. That's just not how it works. If all the good things from Gnome and KDE were in the same Desktop, we would have a better desktop.

>Competition is what fuels these discoveries.

They seemed to have discovered a desktop that is a cross between Windows and OSX. That wasn't a result of competition with each other.

>You're contending Microsoft continues to innovate, so we don't need the ability to experiment on our own, aren't you?

No, I am contending that market forces as a sole force of innovation is not a bad thing. I am not contending that MSes motivations for innovation are solely based on the market however. And I must further contend that Open Source is not the sole force driving MS.

One of MS most innovative research projects of late is MSes fully managed OS initiative. They already have a fully managed kernel called Singularity. Did they make this in response to market forces of other fully managed OSes? Do you know of any fully managed OSes? The closes thing is the JavaOS. But that was dead as a contender before this initiative started and is currently completley dead. Yet, the research in to this new kernel goes on.

>Where do you draw the line then? Is Konqueror still an experimental fork of GtkHTML?

Since it couldn't be, no.

>Since Webkit is merging back into KHTML, is it still a proof-of-concept?

WebKit was never proof-of-concept. All you have pointed out is that a forked project realize being forked was bad. If forking to separate projects was the best course, then they would remain separate. Forking was determined not to be the best course, hence the merging.

>Do you really want to pick such a narrow definition of fork to satisfy that?

Stop supporting my statements and we might be able to have a conversation.

>That's a logical fallacy. If I say webkit (for example), you simply say it wasn't absolutely necessary for Apple to fork, and there's no way to disprove it because it isn't falsable.

Well then provide a more concrete example. The whole falsable thing works in reverse too.

81 Name: #!/usr/bin/anonymous : 2008-09-09 16:08 ID:GJSJH4YG

> Your inability to explain this makes me wonder exactly how much work you did on this project.

You inability to spell or use complete sentences makes me wonder exactly how relevant your opinion is.

> Are you trying to avoid loading paged out virtual memory to physical memory when your CG is running?

I used it to avoid page faults when garbage collecting (GC). Using an incremental collector meant it was always running.

> you could easily infer from my original statement that I was not talking about this type of forking.

I certainly could not! You infer that your inability to understand me is my fault, and that my inability to understand you is my fault.

You were talking about resource sharing. All of these things fork a codebase and separate the theoretical development resources, but from a practical point the resources aren't compatible anyway.

> They (Cedega) took Wine, made a proprietery version of it and refused to contribute back.

Wishing is nice, but it's not exactly relevant. For whatever reason, they didn't contribute back. If they were required to contribute back, they probably wouldn't have bothered making the changes at all.

That doesn't demonstrate resource-contention is occurring, which means it doesn't underscore your point: without the fork, the changes wouldn't have been made at all.

> Wine likes forking to proprietery stuff. Is the best concequence of forking that it can end forking?

Of course it isn't. NeXT forked GCC and back game the objective-C compiler. Sometimes they contribute back, sometimes they don't.

My point is that forking lets them try things out faster.

> It's a safe assumption they bought it to make OSX. They only stated as much back when they bought it that they were purchasing it to make their next OS for Macs.

But it's not a safe assumption when you said "Mac OS X took less time to develop than any Linux desktop" you really meant "The development Apple did covers less time than the development that KDE did at version 1.0" which quite frankly is very silly. I'm surprised you think anyone else could assume that you meant that.

> (KDE is) Still not as good. Go up until now, still not as good.

Says you. I find Mac OS X counterproductive for a laundry-list of reasons- it's certainly harder to use than NeXT.

However, your opinions are your own.

82 Name: #!/usr/bin/anonymous : 2008-09-09 16:17 ID:GJSJH4YG

> Good things happened on the browser (the MS browser) and web during those other 7 years where MS dominated.

You just can't name any of them?

> If all the good things from Gnome and KDE were in the same Desktop, we would have a better desktop.

Agreed. However that's impractical because you'd also be sharing the worst parts of Gnome and KDE in the same desktop.

Just as the GNOME1 corba to GNOME2 dbus/gconf system was a radical shift, KDE3 and KDE4 are radically different systems as well.

This experimentation will be good in the long run because we can learn a lot from these things. GNOME's Dashboard project seemed like such a good idea that Microsoft tried to steal it and run with it- even marketed Longhorn as supporting the model intimately.

However, neither happened. Why? Experimentation demonstrated it sucked in practice- we're not there yet. Maybe fruition will show up, and maybe not. Forcing GNOME or KDE to wait around while we find out means there is less innovation, not more.

> WebKit was never proof-of-concept. All you have pointed out is that a forked project realize being forked was bad. If forking to separate projects was the best course, then they would remain separate. Forking was determined not to be the best course, hence the merging.

In this case, the fork- turns out to be superior and effectively replaces KHTML. If Apple's heavy lifting hadn't turned out as well, WebKit would've been dropped like a stone.

> Well then provide a more concrete example. The whole falsable thing works in reverse too.

Quit being childish. I've demonstrated forking is sometimes good. I don't have to demonstrate that it is best to prove you wrong. You said forking was always bad- I only had to prove that it wasn't, and I have done that.

83 Name: #!/usr/bin/anonymous : 2008-09-09 19:46 ID:uDj+1DBL

>If they were required to contribute back, they probably wouldn't have bothered making the changes at all.

That is an odd thing to say. Probably. Probably the Large Hadron Colider will kill us all. Factually, it will not.

>My point is that forking lets them try things out faster.

If someone want to branch a code base to try something out, I would call that branching not forking. If the intent is to merge the experimental code back in to the project, it really isn't a fork that creates 2 separate projects. The projects are inherently linked and in the end the branch will be discarded or merged. End result, not 2 separate projects. End result, same goals.

>But it's not a safe assumption when you said "Mac OS X took less time to develop than any Linux desktop" you really meant "The development Apple did covers less time than the development that KDE did at version 1.0" which quite frankly is very silly. I'm surprised you think anyone else could assume that you meant that.

The first release of MacOS is still better than any Linux desktop. And I don't need some stupid but Linux does these things better crap. All around, it is a better Desktop OS.

From the beginning of NeXT up until Mac OSX's first release, is still less time than Linux and been in development.

>Says you. I find Mac OS X counterproductive for a laundry-list of reasons- it's certainly harder to use than NeXT.

There is quite a consensus that OSX is still a better desktop than Linux. The market certainly spoke out about it. Many large OEMs have tried making offerings for many years now. Still not as successful as OSX. Individuals have been able to easily acquire it since inception, and still the market share on the desktop is less than 2%.

I am not trying to inject my personal opinions. I am just using same facts as they are today.

>I've demonstrated forking is sometimes good.

You have demonstrated that branching is good. Forking doesn't seem to end up doing to well.

84 Name: #!/usr/bin/anonymous : 2008-09-09 21:03 ID:GJSJH4YG

> If someone want to branch a code base to try something out, I would call that branching not forking. If the intent is to merge the experimental code back in to the project, it really isn't a fork that creates 2 separate projects. The projects are inherently linked and in the end the branch will be discarded or merged. End result, not 2 separate projects. End result, same goals.

So something is a fork if at the time it was made, there is no intention of merging it back in later?

So WebKit is a fork, Linux 2.2 is a fork, but Linux-mm (which is called a fork by the Linux developers) is not a fork?

What a bizarre world you live in.

> The first release of MacOS is still better than any Linux desktop.

In your opinion.

I actually used the first release of Mac OS, and I am of the opinion it is garbage.

You are welcome to your opinion, but calling it something else doesn't make it something else.

> From the beginning of NeXT up until Mac OSX's first release, is still less time than Linux and been in development.

For some narrow and meaningless semantics, certainly. For some equally narrow and meaningless semantics, the inverse is true.

Try using a quantifiable language. Make falsible statements if you're trying to prove something.

> There is quite a consensus that OSX is still a better desktop than Linux.

There is quite a consensus that Linux is a better desktop than OSX. If you can't argue to any affect that doesn't require I share your opinion, you're just being an asshat.

> The market certainly spoke out about it.

By that logic, Windows is a better desktop than Mac OS X, and Internet Explorer 6 is a better web browser than Internet Explorer 7.

Better isn't a quantifiable statement. It speaks of quality, and market sizes are never an indicator of quality. That means the market sizes cannot say anything about whether something is qualitatively better or not.

85 Name: #!/usr/bin/anonymous : 2008-09-11 03:39 ID:Kj0B96tp

>So WebKit is a fork

WebKit was a fork. The consequence being, the KHTML devs realized that forking was bad and want to merge their efforts in to one superior effort.

>Linux 2.2 is a fork

HAHAHA, a Linux kernel fork? No. Sure, anyone can claim Linux fork. But I am talking about a practical viable for of the Linux kernel. Not happening. The at large community doesn't even want such a thing. I wonder why????

>Linux-mm

Is a fork of what, and the devs call it a fork where?

>You are welcome to your opinion, but calling it something else doesn't make it something else

Given the choice between the various OSes, more people are of the opinion that that OSX is better. I use market share as a simple illustration of this. When OSX came out Macs were hurting. They are doing better now, compared to themselves before and compared to Linux, despite the wider viability of Linux. People may pay extra for a Mac because of brand loyalty or marketing, but its hard to argue that they don't also feel Mac OS is a better desktop than Linux.

>Try using a quantifiable language. Make falsible statements if you're trying to prove something.

Mac OS has greater market share and a faster rate of adoption on the desktop given any comparable time frame compared to Linux on the desktop.

>By that logic, Windows is a better desktop than Mac OS X

Well yeah, it is.

>and Internet Explorer 6 is a better web browser than Internet Explorer 7.

Given that IE7's market share is larger (by close to 2x) and its adoption rate faster than IE6's, I would say no and many others agree.

>Better isn't a quantifiable statement.

True, I am speaking of a general consensus in the market and not just a bunch of louder than they need to be idiots (not saying that is you, but if you believe a lot of blogs out there you would believe that every MS program gives you AIDs).

>It speaks of quality, and market sizes are never an indicator of quality.

I would say so if the products are close in accessibility and function. Its easy to say that a BMW 5 series is "better" than a Toyota Camry. Both may be sedans but price makes the BMW less accessible. The Camry sells much more but is not of the same quality. The market doesn't help much there.

But Between Vista, OSX and Linux we have a completley different situation.

Linux is the most accessible with its free price tag (or very low relatively) compared to the others. It runs on a larger number of desktop platforms. It is the most accessible, yet has the smallest market share.

OSX is less accessible because of price and hardware limitations. Yet, it is many times more prevalent in the market. Again, some is brand loyalty, some is marketing but that is not all of it. Some people who have an informed choice between all made a choice. And it is reasonable to infer that number is significant in comparison to the Linux desktop share.

Windows sits in the middle. Runs on more platforms than OSX (including the same newer platform as OSX) but less than Linux, the price is in the middle and yet it is far more prevalent on the desktop. Now of course some of that is because an MS employee came to my house and rammed a gun down my throat and made me run it (oh wait, that didn't happen) but it is far and away the leader by a huge margin.

People do have a choice. Some chose it for the wrong reasons and if the numbers were close we could argue that. But the numbers are not close. They are far enough apart that we need can come to some conclusions. User obliviousness will account for much of that, but not all. And this day and age, more people are computer savvy and aware of Linux than ever. But that isn't helping.

Oh, and have you seen the Linux vs. Windows Server adoption rates over the last 2 years. That shit will blow your mind (hint: Linux ain't doing so well). Can't really say sysadmins aren't aware of Linux or completley removed from its choice of servers. But that really is a different topic from the desktop.

86 Name: #!/usr/bin/anonymous : 2008-09-11 17:42 ID:GJSJH4YG

> Mac OS has greater market share and a faster rate of adoption on the desktop given any comparable time frame compared to Linux on the desktop.

Citation needed.

> Given that IE7's market share is larger (by close to 2x) and its adoption rate faster than IE6's, I would say no

IE7's market share wasn't always larger. Does that mean IE6 was at one point better?

> many others agree.

Citation needed.

> Linux is the most accessible with its free price tag (or very low relatively) compared to the others.

Rejected.

Dell sells Linux PC's at a higher price than it sells Windows PC's, and being as how you can't find Linux PC's in Best Buy, I'd say that the penetrated/effective price approaches Inf.

People are going out of their way to install Linux. I've never seen white-box sales of Microsoft Windows ever approach that of Linux downloads, and I'd be suspicious of any report to the contrary.

> Some people who have an informed choice between all made a choice. And it is reasonable to infer that number is significant in comparison to the Linux desktop share.

Citation needed.

> Oh, and have you seen the Linux vs. Windows Server adoption rates over the last 2 years

According to IDC, Linux penetration saw 12.7% to Microsoft's 10.4%. Meanwhile, netcraft also has a different definition of "ain't doing so well" with Linux continuing to dominate with more than 60% of web servers.

Of course, you don't have to back up anything you say with citations or logic or anything; you haven't thus far, and I won't expect you to now, either.

87 Name: #!/usr/bin/anonymous : 2008-09-11 21:29 ID:Kj0B96tp

>Citation needed.

http://marketshare.hitslink.com/report.aspx?qprid=9&qpdt=1&qpct=4&qpcustom=Mac&qptimeframe=M&qpsp=92&qpnp=25

http://marketshare.hitslink.com/report.aspx?qprid=9&qpdt=1&qpct=4&qpcustom=Linux&qptimeframe=M&qpsp=92&qpnp=25

>IE7's market share wasn't always larger. Does that mean IE6 was at one point better?

That means at one point IE7 was new.

>Citation needed.

http://marketshare.hitslink.com/report.aspx?qprid=3

>Dell sells Linux PC's at a higher price than it sells Windows PC's

Yeah, because Dell is always doing things right. But why should one expect to get a pre-built linux PC for significantly cheaper anyway? Linux may be free, but other things surrounding it (support and such) are not. And Windows OEM licenses are cheap as hell. Also, anyone who knows about this tuff knows that licesnse fees for OSes don't take an enormous bite out of the IT budget. Running Linux doesn't end up being signifgantly cheaper in the long run. TCO is a complicated subject and there is no clear winner. So its easy to say in general Linux is not signifigantly cheaper to operate than Windows.

>and being as how you can't find Linux PC's in Best Buy

You can buy a bunch of different distros of Linux at Best Buy. Been that way for something like 10 years or more.

How about we look at the largest retailer in the US.

http://blog.wired.com/business/2008/03/middle-america.html

Oh shit, Linux failed.

So the largest PC retailer sells Linux PCs. The largest consumer electronics store sells Linux. The largest retailer sold Linux PCs. Anyone can download Linux for free. You can try it without installing anything.

Linux is out there, it is available, it is accessible, people know about, people have tried it.

>People are going out of their way to install Linux. I've never seen white-box sales of Microsoft Windows ever approach that of Linux downloads, and I'd be suspicious of any report to the contrary.

Which of course is curious because Windows is still on more PCs. So using downloads a guide of adoption doesn't make sense. Constant releases mean more downloads. People trying out a few distros means more downloads. Not every download comes close to an installed and regularly used Linux implementation. I have had 6 Windows versions I have regularly used and I have downloaded close to 30 Linux distros over that same time period.

>Of course, you don't have to back up anything you say with citations or logic or anything; you haven't thus far, and I won't expect you to now, either.

Speaking of logic, I was talking about recent adoption rates and not overall numbers limited to 1 aspect.

Lets look at 2006-2007:

"According to research firm Gartner, the Windows share of global server shipments gained a percentage point to 66.8 percent in 2007 from a year earlier. Open-source Linux's share fell by a percentage point to 23.2 percent last year and Unix dropped to 6.8 percent in 2007 from 8.1 percent in 2006."

http://www.reuters.com/article/ousivMolt/idUSN2748543820080228

So you can limit your focus to just web servers and broaden your time frame to avoid the point.

I'll just show you that over the last few years Windows Server is kicking Linuxes ass in the server market.

88 Name: #!/usr/bin/anonymous : 2008-09-12 12:21 ID:Heaven

>>87

Oh, eighty-seven, you are a precious child.

http://marketshare.hitslink.com/ measures their information by looking at their advertisement logs. I cannot see any reason to believe that they are representative, you failure.

The Gartner quotes server shipments, it doesn't quote installed-base, and has no way to do so. Random surveys demonstrate the opposite trends, which indicates that Gartner is wrong, and you are a failure for not knowing how to read.

Finally, Windows Server 2008 with five licenses is almost a thousand dollars. That doesn't include any support, nor any training. I suppose you can call whatever costs you like "insignificant", but it doesn't excuse the fact anyone else will look at you and see a failure.

I don't even know what you're arguing anymore. Microsoft comes into my shop every few months to tell me the same thing you're telling me, and try and sell me new licenses. I show them my books, and they leave heads hung low.

I rejected your premise that forking was always bad, and now you're trying to convince me that Linux is a market failure because Gartner- a company who frequently takes money from Microsoft directly- says it is? Because Wal-mart's choice to drop Linux from one retail outlet to zero somehow demonstrates that tracking retail sales of Linux is the same as tracking Linux?

You've failed at everything here; at logic, at a point, and at convincing anyone of anything.

89 Name: #!/usr/bin/anonymous : 2008-09-12 21:29 ID:RckJZNrp

>http://marketshare.hitslink.com/ measures their information by looking at their advertisement logs. I cannot see any reason to believe that they are representative, you failure.

Ahhh the easiest Internet debate technique. Request source then reject it. Please do name some acceptable sources.

>The Gartner quotes server shipments, it doesn't quote installed-base, and has no way to do so.

I can imagine a significant number of shipped server never get used /sarcasm

>Random surveys demonstrate the opposite trends, which indicates that Gartner is wrong, and you are a failure for not knowing how to read.

Magical un-cited surveys.

>Finally, Windows Server 2008 with five licenses is almost a thousand dollars.

Fair enough.

>That doesn't include any support, nor any training.

Linux also shares in those types of costs so they don't really factor in to a the comparison.

>I suppose you can call whatever costs you like "insignificant", but it doesn't excuse the fact anyone else will look at you and see a failure.

Let's go with MSes "Mainstream" support as a basis for figuring some costs for a server. That is currently 5 years. Over 5 years $1,000 isn't that significant when you factor in hardware costs, 5 years of admin staff (even if it is just 1 person for 5 servers), utility bills etc. That $1,000 doesn't even cover the cost of a competent admin for a week.

>I don't even know what you're arguing anymore. Microsoft comes into my shop every few months to tell me the same thing you're telling me, and try and sell me new licenses. I show them my books, and they leave heads hung low.

I am not telling you Linux isn't good at anything. It works for some organizations. Windows Server works better for more.

I am curious as to what database solution you are using if any.

>I rejected your premise that forking was always bad

And you somehow did that by point to projects that have either unforked, or the forks clearly demonstrate some fundamental issues of forking.

>and now you're trying to convince me that Linux is a market failure because Gartner- a company who frequently takes money from Microsoft directly- says it is?

Show me a better source that isn't reflective of just one convenient portion of the market.

>Because Wal-mart's choice to drop Linux from one retail outlet to zero somehow demonstrates that tracking retail sales of Linux is the same as tracking Linux?

Because it demonstrates Linux's current in-viability as a desktop OS.

>You've failed at everything here; at logic, at a point, and at convincing anyone of anything.

Yet you present counterclaims with no demonstrable basis.

90 Name: #!/usr/bin/anonymous : 2008-09-16 22:23 ID:Heaven

> Ahhh the easiest Internet debate technique. Request source then reject it. Please do name some acceptable sources.

No, the easiest internet debate technique is to make a vacuous claim like "forking is always wrong" and refuse to back it up with facts or arguments.

Remember, you just said Microsoft is demonstrating growth in the browser market because hitslink.com's logs say so. You have to demonstrate that their sample set is representative.

> Over 5 years $1,000 isn't that significant when you factor in hardware costs, 5 years of admin staff (even if it is just 1 person for 5 servers), utility bills etc. That $1,000 doesn't even cover the cost of a competent admin for a week.

Unfortunately, you still need the hardware, and you still need the admin staff, and you still need the utility bills.

I've got 60 servers in my server room right now. You're suggesting an overhead of 12,000$ on my small site alone, and to what end? There's absolutely no way I could sell that to anybody.

> I am curious as to what database solution you are using if any.

It varies. Pg for most things.

> And you somehow did that by point to projects that have either unforked, or the forks clearly demonstrate some fundamental issues of forking.

Balderdash. I pointed to webkit and cedega as obvious examples.

You still have to demonstrate that if forking Cedega were impossible, the resources could still have gone to making Wine better, in order to reject it.

You would also need to do something similar for webkit.

I don't think it's possible to demonstrate either of those, which is why I think you're wrong.

> Show me a better source that isn't reflective of just one convenient portion of the market.

I don't think that's possible. Firstly, I still reject your notion that market penetration is a measurement of "better" because you cannot reject the null hypothesis- that it doesn't have penetration for some other reason.

Secondly, this argument is effectively "competition is bad", and yet I don't see any reason to believe that.

> > You've failed at everything here; at logic, at a point, and at convincing anyone of anything.
> Yet you present counterclaims with no demonstrable basis.

I don't have to. You made the bizarre claim; the onus is on you to back it up. You just admitted that you have used no logic in your broad statements of forking always being bad, and have no particular points to back it up.

91 Name: #!/usr/bin/anonymous : 2008-09-17 19:41 ID:OBI63/NB

>No, the easiest internet debate technique is to make a vacuous claim like "forking is always wrong" and refuse to back it up with facts or arguments.

You did that by pointing out forks that either unforked, were branches instead of forks or the developers changed something to make it so they couldn't fork it.

>Remember, you just said Microsoft is demonstrating growth in the browser market

See how easily you get confused. I said IE7's market share is growing relative to IE6, not overall. I said that MSes server OS share is growing relative to Linux.

>I've got 60 servers in my server room right now. You're suggesting an overhead of 12,000$ on my small site alone, and to what end? There's absolutely no way I could sell that to anybody.

That overhead may not be worth it to you. Linux is viable for you. However, those license fees over their mainstream life are insignificant compared to all of the other costs associated with operating the servers during that life. And I am saying insignificant because as you know, $12,000 doesn't go a long way in 12 years with 60 servers.

>It varies. Pg for most things.

And it's a fine choice for many implementations. But it still doesn't meet the needs of everything. Just like MySQL is good for some websites but not others. If you need something more robust, many go for a commercial solution. And if you aren't doing Oracle (or DB2 etc) the most likely choice is going to be MS SQL Server. And those solutions still cost more than the server OS license.

My point being, Windows is gaining share because even though Linux has a lot of options and can do a lot, Windows Server can do a lot too and some of the Windows only offerings are superior to what is free on Linux, or if you go with a commercial Linux solution the cost typically negates the cost of a Windows Server license.

>You still have to demonstrate that if forking Cedega were impossible, the resources could still have gone to making Wine better, in order to reject it.

What? TransGaming is the only one that can fork Cedega. I think you mean to say that if Cedega contributed back to Wine it would be better. There isn't a concrete answer to that. But you can infer from the fact that the Wine developers changed the Wine license to prevent such a fork from being created again, that the developers found that fork to be detrimental to their goals.

>You would also need to do something similar for webkit.

If forking was working so well, it will still be forked.

>I don't think that's possible. Firstly, I still reject your notion that market penetration is a measurement of "better" because you cannot reject the null hypothesis- that it doesn't have penetration for some other reason.

It is easy to infer that Windows Server has been found to be a "better" solution to more tasks than Linux for more server admins.

>Secondly, this argument is effectively "competition is bad", and yet I don't see any reason to believe that.

I am not saying that. If anything Linux does drive MS to do things to an extent, and that's good. But MS seems to always stay ahead of Linux despite companies like IBM pushing it.

>You just admitted that you have used no logic in your broad statements of forking always being bad, and have no particular points to back it up.

The logic would be if forking is good, good things would come from it.

The Wine devs felt the Cedga fork was bad. The CrossOver "fork" is made by the Wine devs. Wine is the same in this fork, the only thing it adds are proprietary support tools so it functions more like the commercial branch of Wine.

WebKit is unforked. If forking was good, then it would still be forked.

92 Name: #!/usr/bin/anonymous : 2008-09-18 04:02 ID:VIUDwxgD

> I said that MSes server OS share is growing relative to Linux.

And after some arm twisting, we find you're quoting The Gartner group. Meanwhile, IDC's numbers disagree.

It's obviously not clear who's share of what is growing because there are so many different things to measure here.

> And I am saying insignificant because as you know, $12,000 doesn't go a long way in 12 years with 60 servers.

Windows NT 4 would be 12 years old. Are you still running it?

> many go for a commercial (DB) solution

We go for a commercial solution when we want support. Microsoft's support is garbage, ergo, we don't buy Microsoft.

When we ran SAP, they gave us source code, and gave us solutions. Not all vendors are created equal, and when it comes to practical solutions, Pg offers better performance and better support than most commercial setups.

> that the (wine) developers found that fork to be detrimental to their goals.

I agreed with that. They felt cheated that someone could take their hard work and commercialize it. Yet another reason to distrust commercial packages without source code.

However, you'll note there haven't been any forks since- even though forks are still possible. Were the Cedega fork impossible, we probably wouldn't see Cedega in any form. We most certainly wouldn't see the changes that Cedega saw either simply because the resources that made Cedega were incompatible with the resources that are devoted to Wine.

> > you cannot [prove] that [windows] doesn't have penetration for some other reason.
> It is easy to infer that Windows Server has been found to be a "better" solution to more tasks than Linux for more server admins.

Sure. You can infer that Purple Unicorns eat puppies. If you don't have any evidence to back it up, you don't have any evidence to back it up.

> I am not saying that. If anything Linux does drive MS to do things to an extent, and that's good. But MS seems to always stay ahead of Linux despite companies like IBM pushing it.

That's a different debate. If this one is any indication, I don't think you're up for it.

> The logic would be if forking is good, good things would come from it.

Good things can come from it.

> The Wine devs felt the Cedga fork was bad.

The Cedega people felt that the Cedega fork was good. People who purchased Cedega thought it was good as well.

> WebKit is unforked. If forking was good, then it would still be forked.

Not necessarily. I maintain the act of forking doesn't have to be permanent because it wasn't in your original definition- only that the intent were that the projects not re-merge. Webkit wasn't forked under the intention that it re-merge.

93 Name: #!/usr/bin/anonymous : 2008-09-19 07:09 ID:OBI63/NB

>It's obviously not clear who's share of what is growing because there are so many different things to measure here.

Yeah, because measuring the number of OSes servers are running is too inconvenient.

>Windows NT 4 would be 12 years old. Are you still running it?

Obvious typo since I was talking about the 5 year mainstream life-cycle up until that. Point still stands at 5 years.

>Not all vendors are created equal, and when it comes to practical solutions, Pg offers better performance and better support than most commercial setups.

Tis true. Performance is a good reasons to go with pg. Support? You mean relying on the community (which you can do equally as well with MS products), or hiring a consulting firm (which you can do equally as well with MS products)? Those are the only 2 options for support with pg.

But when reliability and features are absolutely required, commercial is the way to go. MS SQL has so many features and mature tools that free database solutions just do not offer.

>Were the Cedega fork impossible, we probably wouldn't see Cedega in any form. We most certainly wouldn't see the changes that Cedega saw either simply because the resources that made Cedega were incompatible with the resources that are devoted to Wine.

Many feel that the changes that the Wine devs made to match Cedega actually surpass it. And it is not like it was a competition, those changes were always inline with the Wine road map.

>If you don't have any evidence to back it up, you don't have any evidence to back it up.

Higher and rising market share at a faster rate than the competition.

>Good things can come from it.

Have there been concrete things that are absolutely great and required forking to accomplish?

>People who purchased Cedega thought it was good as well.

Until they didn't and they had to focus on proprietary solutions for Mac. Not a bad move on their part, but not a move that absolutely required Cedega to go the route they did.

>only that the intent were that the projects not re-merge. Webkit wasn't forked under the intention that it re-merge.

Very true.That intention was found to be an error.

94 Name: #!/usr/bin/anonymous : 2008-09-19 19:05 ID:GJSJH4YG

> Yeah, because measuring the number of OSes servers are running is too inconvenient.

Do that.

Go randomly sample a million sites. It's not hard. Get a list of IP addresses, do PTR lookups on a million at random, and pipe it into nmap. It should take you about three to five days on a fast connection.

Instead, you're measuring the number of servers sold in a given marketplace. That's a measurement of purchase trends, not on system penetration.

> Obvious typo since I was talking about the 5 year mainstream life-cycle up until that. Point still stands at 5 years.

Not obvious typo. I have several UNIX systems that have uptimes over a decade.

Never mind the fact that $12,000 when compared to zero, is enormous.

> Tis true. Performance is a good reasons to go with pg. Support? You mean relying on the community (which you can do equally as well with MS products), or hiring a consulting firm (which you can do equally as well with MS products)? Those are the only 2 options for support with pg.

Obviously, you've never hired an MS consultant.

Here are some wonderful things I hear from Microsoft's support staff:

  • It looks like you should reinstall and restore from backup.
  • You can't run Exchange on the same machine as your domain controller
  • You can't run Raiser's Edge on the same machine as Exchange
  • You can't run SQL Server Express on the same machine as SQL Server without massive performance penalties
  • Using Dynamic Disks and mirroring protects you against drive failure

The fact is, all consultants are idiots. Some are more so than others.

However, in an open market, those idiots can't as easily survive. As a result, the Pg consultants (who must be good in order to compete with each other) are far superior to the Microsoft consultants (who don't have to compete at all).

> Have there been concrete things that are absolutely great and required forking to accomplish?

I never said there were. I said there were things that turned out absolutely great, that were the result of forking.

Stop creating a straw man. The arguments are simple:

  1. Webkit is the result of a fork, and webkit is good ergo forking is sometimes good.
  2. Cedega is the result of a fork made up from a codebase that wouldn't exist were the fork impossible (legally), ergo there is no contention of resources.

That means you're wrong. Quit with the hand-waving and deal with it directly.

95 Name: #!/usr/bin/anonymous : 2008-09-24 21:43 ID:Heaven

You guys are so intelligent you'll derail a thread to have an argument. About how much more intelligent you are than the other guys doing the same stupid thing.

For shame.

96 Name: #!/usr/bin/anonymous : 2011-01-25 16:08 ID:Wt2vPB6b

sir database is not working. there is a problem with the "ban" table

97 Name: #!/usr/bin/anonymous : 2011-01-25 20:10 ID:Heaven

Oh hey, I remember this thread. It was stupid.
Thanks >>96 for letting us relive stupid moments of internet history.

98 Name: AcompleteNOOB-jk : 2011-11-13 20:41 ID:d4ZlCwEu

Okay here is some software that supposedly does imageboard, forum and cms all rolled into one, they went offline for like 2 days but their back check it..

www.rateitstar.com/software.php

some guy who calls himself Ere wrote it, lol, but the software is some serious hot shit..

99 Name: #!/usr/bin/anonymous : 2011-11-14 05:50 ID:OxEDdlIY

>>98
I'd be embarrassed if my name was associated with something that bad.

100 Name: AcompleteNOOB-jk : 2011-12-29 03:15 ID:mc6O8Y7d

lol

101 Name: #!/usr/bin/anonymous : 2012-06-17 19:29 ID:uXTwXzlw

102 Name: #!/usr/bin/anonymous : 2012-07-20 21:22 ID:93/NkD8C

Lets talk about the state of things in the imageboard script world today.

Is it true TinyIB was another attempt at making an imageboard script by the guy who made Trevorchan, Kusaba, etc.?

Is TinyBoard's development halted forever? Did anyone ever find out where it stands compared to Wakaba in terms of both speed and security?

Has anyone found any unpatched security holes in Wakaba yet?

What exactly happened to STI?

Are there any new contenders (not forks of other scripts) in the great imageboard script race?

103 Name: #!/usr/bin/anonymous : 2012-07-22 21:12 ID:Heaven

104 Name: #!/usr/bin/anonymous : 2012-08-06 15:18 ID:93/NkD8C

>>103
Jesus fuck I'd love to see this thing in action.

105 Name: #!/usr/bin/anonymous : 2012-08-06 15:18 ID:Heaven

106 Name: #!/usr/bin/anonymous : 2013-11-23 13:54 ID:C/qJk7Ke

107 Post deleted.

108 Post deleted.

109 Post deleted.

110 Post deleted.

111 Post deleted.

112 Post deleted.

113 Post deleted.

114 Post deleted.

115 Post deleted.

116 Post deleted.

Name: Link:
Leave these fields empty (spam trap):
More options...
Verification: