Okay, you're right. Unlimited installs are not a "feature" of Steam, since all other models inherently have that feature already. You have fun digging around for and swapping out install discs while I go play any game I want on this newly built PC.
And what if I want to transfer the same encrypted / signed message via sneakernet to someone across the hall? I find that most "all inclusive solutions" have the attitude of "this solution is all-inclusive, so of course you wouldn't ever want to do anything involving anything other than what this covers."
Sometimes you just want to hand a courier* a thumbdrive with a file on it without worrying about where he might leave it while he's on lunch
*courier, kur--r (noun): some guy who will pass by steve on his way to a meeting later today
Every Star Trek iteration has been an entirely new crew, ship, and usually mission statement. The only way to not just be "more of the same" is to revisit existing characters.
I tried to start using encryption once, so I installed what I thought was an encryption program.
Turns out what I installed was some sort of huge "key management system" with all sorts of ins and outs and a whole epic mountain of a learning curve which I was expected to either:
- Learn all of intimately before I began or
- Ignore and pretend I was being a neato-l33to secure communications boy just because some random program I had installed told me I might be.
I don't need to understand the math behind it, but I do need to understand the concepts of what is going on. I should be able to say "Oh, I just received an e-mail, try using this key." instead of telling a huge system, which I presumably need to set up on every computer I ever touch, what to do for me.
No, spammers don't care enough about that particular website to crack it. If your website is a spam target, you will be spammed. Otherwise, you can just have a checkbox that says "check here if you are not a spam bot" and it will provide just as much security.
More or Less fundamentally, depending on how you look at it: Copyright exists to allow authors a monopoly for a LIMITED TIME. Because DRM systems will exist (whether or not they have included measures to somehow disable themselves) after that "LIMITED" time, any work which is "protected" by DRM should not be protected by any copyright law (as it never allows entrance into public domain)
Of course, the same applies to all closed-source software- if you don't release source code, and therefor do not give works a chance to ever enter the public domain in a reasonable fashion, your compiled programs should not be protected.
You can't get a patent on an invention without telling everybody how it works, you shouldn't automatically get "copyright" on something when you have gone out of your way to make sure people cannot usefully copy it once the copyright expires.
wouldn't want any song I own placed in this movie...
What makes you think your wishes about how you want something to be used matter? Oh right, broken and horribly mis-applied copyright laws.
Use without paying royalties is one thing, but this "moralistic" argument that artists have a right to "control their work" is bullshit and has no place outside of informal politeness. (read: Not courtrooms)
When circumvention is used to transfer content from one device* to another, (or from one owner to another in such a way that the original user no longer has access to the content [such as uploading a file, then deleting the local copy]), in a circumstance which would not normally be permissible due to technical, but not legal, restrictions of the scheme being circumvented.
*"device" is a poor choice of phrase, as this should also cover use of other operating systems or players on a single physical "device".
no, no, don't get me wrong. LSL, the language, sucks. I tried to make that clear in my post, but I guess I failed it. "Even ignoring the uselessness of the built-in functions it can't be seen as good." was meant to refer mostly to the API. Obviously it's a bit hard to separate what is part of the "API" and what is part of the basic language (at least when talking about list manipulation), but that's probably one of the problems with the language itself.
The whole thing is bulky, clunky, and a horrible pain to use. Perhaps Mono is making it faster, I've been warned away from the latest RC so haven't tried compiling to it yet. I somehow doubt that Mono is magical, though, and expect that it's quite possible to make things slow and broken even in a compiled byte-code (O(n) is still O(n), no matter how it's represented).
When I say the "basic structure" I mean I have nothing against:
But the way variables are handled, particularly lists, is rather non-sane. The built-in functions for variable manipulation are very weak. It's difficult to draw the line between the "basic functions" and the "SL Library". You could call this favoring consistency, but come on, this isn't Lisp, saying "it's all functions" doesn't work towards beauty.
[note: I won't have a chance to try the new version until later tonight, so this post is based on earlier versions. This sounds fair to me, since the "MDI/SDI" debate has raged for centuries, and has, until now at least, been completely inapplicable to Gimp, which is neither]
Got it. Agreed. Now show me ANY window manager which handles such a thing as well as, say, Photoshop's MDI for a single application.
It seems that everyone who makes this argument seems to be of the "virtual desktop" bunch (usually 1 application per desktop). ie: Those who don't actually use the primary feature of a windowed environment: Windows!
Meanwhile, GIMP tries to have it both ways, sharing arbitrary windows whose context depends on the last window selected, while (arbitrarily) putting "cross-window" features in [not a shared interface, but instead:] EVERY WINDOW.
There's MDI, there's SDI, and then there's GIMP, which has taken the worst features of both.
When I get home, I'll download the latest version, which may have actually addressed all of these complaints (the release notes tease more than any previous versions' has. I still expect it to be usability hell; but, for example, removal of the menu bar from the toolbox window sounds very promising.
LSLs horrendous nature has nothing to do with it being a "real time control system". It's just bad.
The types of very-domain-specific languages you speak of are that way because of their efficiency. I don't think you or anyone else would accuse LSL of being efficient. Can you even design in your head right now a system which would work in such a way that removing (perhaps 200 total, across a sim) attachments which have set up a "listen" could reduce lag a non-trivial amount?
LSL's basic structure isn't bad. It's nice enough that I really think someone would use it (another "Real time" system counts) if it weren't horrible and inconsistent. Even ignoring the uselessness of the built-in functions it can't be seen as good. (Maybe Mono will bring the age of libraries.. dream dream dream)
Google has _no_ content creation now? Is that really that much of a step down from what Second Life supports?
Infinitely less.
Well, sure, that's true if you go by what the words mean:)
"In-world creation of content" == "In world _ONLY_"
SL, for something so mature, is in a very sad state as a platform goes.
- No way to legitimately back-up created items (and the unsupported ways of backing up are so universally shunned you can't even mention using them for your own content)
- Very poor offline tools for content creation. Using photoshop and in-world tools some very good designers have made some very awesome stuff. It's even more impressive when you realize that only the most limited tools exist for previewing content you've created. Your options are pretty much: 1) use very bad approximations of the models which support absolutely none of the features of the full viewer or 2) load up the full viewer, import textures manually, manually apply them to all involved objects.
- The built-in scripting language (which is the only option available for interactive content in-world) is horrible. It's almost unheard of that a programming language comes along that some crazy person doesn't swear by and decide to write everything they do in it. But this doesn't happen with LSL- the only external interpreter for LSL is a feature-unrich debugger.
Call me crazy, I know that Lively is a horrible peice of shit, but somehow I think that Google can do better than Second Life in these respects. Google has _no_ content creation now? Is that really that much of a step down from what Second Life supports? Artists will always be able to make amazing things no matter how little they have to work with. I'm betting that Google will at least make it less painful for them./me hits post and goes back to the SL fashion show he's watching
An unadvertised development release which you need to download a separate program to "update" to is _not_ a release in any sane sense of the word. Everyone I personally know who has used Chrome has managed to (with normal browser usage) crash it, fully, within minutes of first installing it.
The point is simple: They said it was designed in such a way that it couldn't be fully crashed with normal usage, and yet it very easily can be. They released it to the whole world in this state. When reports started coming in, they didn't have a fix up for everyone same-day, same-week, or same-month. I haven't seen any press releases from google noting "woops, guess we were completely wrong with all this "processes!!!" stuff. We'll be re-designing the thing so that we're not completely talking out of our asses, and get back to you."
Say you're working on YourCompanyProduct and you release a major new feature to all of your customers at once. Fifteen seconds later you find out it actually doesn't work for many of them. Do you:
a) Put up a message visible to (at least) the effected users, and pull an all-nighter making sure you get it right.
b) Continue developing at your normal pace, telling no one anything and maybe fixing the bugs at the same time as your next feature release.
as a "cowboy coder" (for the most part), I'd like to pipe in: ignore everything he just said.
The only way you will ever be any good at programming is by teaching yourself, writing code in your free time, learning everything you can outside of a classroom. AND outside of work.
Too many people think that programming is something they can pick up by taking a class. I've met the results and it's not pretty.
More importantly: You do NOT need to understand the implications of any low-level operation, because as a Jr. Programmer you should not be the one making those decisions. Premature optimization leads to bad code, and too many new programmers focus on things which they are not qualified to write, or have no business writing. (I have yet to encounter a real-world situation when I have needed to write my own sorting function, hash table, or non-basic tree structure)
Always remember that for all things, someone has already written exactly what you are trying to write. Simply because it already exists, it has had more time than what you are writing has had to work out the bugs. For the love of god, don't write your own "database abstraction layer". Always assume that for any task, someone else knows how to do it better- and collaborate with other people constantly to find out how that is.
People who "aren't trained in programming", but program in their spare time are ALWAYS going to be better at writing code than people who "are trained in programming", but have never written a line of code outside of school or work.
I guess you're talking about projecting into the pupil from some nearby device, not from glasses.
Nope.
Why would strapping it to your face remove the necessity for head tracking? It would likely make the issue even more prominant since at that scale, eye motion is also very important. 3d on your TV, your perspective doesn't change that much when looking slightly to the left (at least, it's rather unreasonable to expect it to). Meanwhile: ever try looking to the left without moving your head? It's a neat trick that humans can do and cats can't. Your perspective completely changes, which you might be able to note by printing out something wide that looks to be in the correct perspective when looking straight ahead, then look to the side. Yes, even without moving your head, your perspective changes just as if you turned your head.
And eyes move a lot faster than heads, too. Processors in a Wii can keep up with slow head motion, but nothing that fits into a pair of glasses will keep up with your eyes darting around. They do that.
If these were simple problems to solve, VR helmets wouldn't have sucked so much.
Chrome's been out for nearly a month now and I don't see any new release any time "soon". With such a poor release, I expected new versions to come out the same day yet here we are, weeks later, and no sign that the problems are even on Google's radar.
If I pushed a product to millions of users by linking to it from the front page of the world's most popular website, saying it was "uncrashable", and then it turned out within minutes of real-world uses that no, it's just as easy to crash as any other browser (I've yet to see a "sad tab"), or any of the other major problems, etc- I'd work towards fixing them ASAP. Where is the new release? Where is the new alpha?
Google fucked up. Forking might wake them up. All good forks get merged in the end, anyway.
Yeah, damn modern programmers always planning ahead and detecting serious flaws, brining the whole system down if needed when something unrecoverable happens. When there's a problem that means all your data is worthless, systems should just stay up and keep right on outputting those worthless numbers! Five years later, when somebody notices, you can apply a patch that makes no sense to anyone and isn't actually accurate ("better than nothing, right?")
and in processing technology to track eye motion on that scale.
Wearing a TV on your face and requiring you to stare straight ahead is NOT an improvement. With the clunkiness of the Wiimote (fun, maybe, but accurate it aint) I expect "head tracking" would be pretty much removed from the equation if you took away the [central point of focus] too.
Slashdot Mirror
Okay, you're right. Unlimited installs are not a "feature" of Steam, since all other models inherently have that feature already.
You have fun digging around for and swapping out install discs while I go play any game I want on this newly built PC.
And what if I want to transfer the same encrypted / signed message via sneakernet to someone across the hall? I find that most "all inclusive solutions" have the attitude of "this solution is all-inclusive, so of course you wouldn't ever want to do anything involving anything other than what this covers."
Sometimes you just want to hand a courier* a thumbdrive with a file on it without worrying about where he might leave it while he's on lunch
*courier, kur--r (noun): some guy who will pass by steve on his way to a meeting later today
Every Star Trek iteration has been an entirely new crew, ship, and usually mission statement. The only way to not just be "more of the same" is to revisit existing characters.
I'm tellin' ya, I almost made a release that was THIS stable!
I tried to start using encryption once, so I installed what I thought was an encryption program.
Turns out what I installed was some sort of huge "key management system" with all sorts of ins and outs and a whole epic mountain of a learning curve which I was expected to either:
- Learn all of intimately before I began
or
- Ignore and pretend I was being a neato-l33to secure communications boy just because some random program I had installed told me I might be.
I don't need to understand the math behind it, but I do need to understand the concepts of what is going on. I should be able to say "Oh, I just received an e-mail, try using this key." instead of telling a huge system, which I presumably need to set up on every computer I ever touch, what to do for me.
-- Someone who uses sftp when he cares
Your radical ideas about deniable encryption have been suggested before.. and actually work quite well! Here ya' go
glad to know they'll magically filter out my packets, as I'm not yet a citizen
No, spammers don't care enough about that particular website to crack it.
If your website is a spam target, you will be spammed. Otherwise, you can just have a checkbox that says "check here if you are not a spam bot" and it will provide just as much security.
More or Less fundamentally, depending on how you look at it: Copyright exists to allow authors a monopoly for a LIMITED TIME. Because DRM systems will exist (whether or not they have included measures to somehow disable themselves) after that "LIMITED" time, any work which is "protected" by DRM should not be protected by any copyright law (as it never allows entrance into public domain)
Of course, the same applies to all closed-source software- if you don't release source code, and therefor do not give works a chance to ever enter the public domain in a reasonable fashion, your compiled programs should not be protected.
You can't get a patent on an invention without telling everybody how it works, you shouldn't automatically get "copyright" on something when you have gone out of your way to make sure people cannot usefully copy it once the copyright expires.
wouldn't want any song I own placed in this movie...
What makes you think your wishes about how you want something to be used matter?
Oh right, broken and horribly mis-applied copyright laws.
Use without paying royalties is one thing, but this "moralistic" argument that artists have a right to "control their work" is bullshit and has no place outside of informal politeness. (read: Not courtrooms)
When circumvention is used to transfer content from one device* to another, (or from one owner to another in such a way that the original user no longer has access to the content [such as uploading a file, then deleting the local copy]), in a circumstance which would not normally be permissible due to technical, but not legal, restrictions of the scheme being circumvented.
*"device" is a poor choice of phrase, as this should also cover use of other operating systems or players on a single physical "device".
This is what is already happening, at the exact rate that we can come up with new tests.
This rate is of course much slower than the rate at which spammers can crack them.
The problem with the word "rotating" is that it implies re-use. Once cracked, the test is worthless forever, not just for a couple of page loads.
You don't know what you're talking about. Please do not talk about such things in the future. Thank you, and have a nice day.
no, no, don't get me wrong. LSL, the language, sucks.
I tried to make that clear in my post, but I guess I failed it.
"Even ignoring the uselessness of the built-in functions it can't be seen as good." was meant to refer mostly to the API. Obviously it's a bit hard to separate what is part of the "API" and what is part of the basic language (at least when talking about list manipulation), but that's probably one of the problems with the language itself.
The whole thing is bulky, clunky, and a horrible pain to use. Perhaps Mono is making it faster, I've been warned away from the latest RC so haven't tried compiling to it yet. I somehow doubt that Mono is magical, though, and expect that it's quite possible to make things slow and broken even in a compiled byte-code (O(n) is still O(n), no matter how it's represented).
When I say the "basic structure" I mean I have nothing against:
But the way variables are handled, particularly lists, is rather non-sane. The built-in functions for variable manipulation are very weak. It's difficult to draw the line between the "basic functions" and the "SL Library". You could call this favoring consistency, but come on, this isn't Lisp, saying "it's all functions" doesn't work towards beauty.
rabble rabble rabble
UPDATE: The Gimp still sucks and addresses none of the issues I've mentioned.
[note: I won't have a chance to try the new version until later tonight, so this post is based on earlier versions. This sounds fair to me, since the "MDI/SDI" debate has raged for centuries, and has, until now at least, been completely inapplicable to Gimp, which is neither]
Got it. Agreed.
Now show me ANY window manager which handles such a thing as well as, say, Photoshop's MDI for a single application.
It seems that everyone who makes this argument seems to be of the "virtual desktop" bunch (usually 1 application per desktop). ie: Those who don't actually use the primary feature of a windowed environment: Windows!
Meanwhile, GIMP tries to have it both ways, sharing arbitrary windows whose context depends on the last window selected, while (arbitrarily) putting "cross-window" features in [not a shared interface, but instead:] EVERY WINDOW.
There's MDI, there's SDI, and then there's GIMP, which has taken the worst features of both.
When I get home, I'll download the latest version, which may have actually addressed all of these complaints (the release notes tease more than any previous versions' has. I still expect it to be usability hell; but, for example, removal of the menu bar from the toolbox window sounds very promising.
LSLs horrendous nature has nothing to do with it being a "real time control system". It's just bad.
The types of very-domain-specific languages you speak of are that way because of their efficiency. I don't think you or anyone else would accuse LSL of being efficient. Can you even design in your head right now a system which would work in such a way that removing (perhaps 200 total, across a sim) attachments which have set up a "listen" could reduce lag a non-trivial amount?
LSL's basic structure isn't bad. It's nice enough that I really think someone would use it (another "Real time" system counts) if it weren't horrible and inconsistent. Even ignoring the uselessness of the built-in functions it can't be seen as good. (Maybe Mono will bring the age of libraries.. dream dream dream)
Well, sure, that's true if you go by what the words mean :)
"In-world creation of content" == "In world _ONLY_"
SL, for something so mature, is in a very sad state as a platform goes.
- No way to legitimately back-up created items (and the unsupported ways of backing up are so universally shunned you can't even mention using them for your own content)
- Very poor offline tools for content creation. Using photoshop and in-world tools some very good designers have made some very awesome stuff. It's even more impressive when you realize that only the most limited tools exist for previewing content you've created. Your options are pretty much: 1) use very bad approximations of the models which support absolutely none of the features of the full viewer or 2) load up the full viewer, import textures manually, manually apply them to all involved objects.
- The built-in scripting language (which is the only option available for interactive content in-world) is horrible. It's almost unheard of that a programming language comes along that some crazy person doesn't swear by and decide to write everything they do in it. But this doesn't happen with LSL- the only external interpreter for LSL is a feature-unrich debugger.
Call me crazy, I know that Lively is a horrible peice of shit, but somehow I think that Google can do better than Second Life in these respects. Google has _no_ content creation now? Is that really that much of a step down from what Second Life supports? /me hits post and goes back to the SL fashion show he's watching
Artists will always be able to make amazing things no matter how little they have to work with. I'm betting that Google will at least make it less painful for them.
Phase 1: Pump atmosphere into space.
An unadvertised development release which you need to download a separate program to "update" to is _not_ a release in any sane sense of the word. Everyone I personally know who has used Chrome has managed to (with normal browser usage) crash it, fully, within minutes of first installing it.
The point is simple: They said it was designed in such a way that it couldn't be fully crashed with normal usage, and yet it very easily can be. They released it to the whole world in this state. When reports started coming in, they didn't have a fix up for everyone same-day, same-week, or same-month. I haven't seen any press releases from google noting "woops, guess we were completely wrong with all this "processes!!!" stuff. We'll be re-designing the thing so that we're not completely talking out of our asses, and get back to you."
Say you're working on YourCompanyProduct and you release a major new feature to all of your customers at once. Fifteen seconds later you find out it actually doesn't work for many of them. Do you:
a) Put up a message visible to (at least) the effected users, and pull an all-nighter making sure you get it right.
b) Continue developing at your normal pace, telling no one anything and maybe fixing the bugs at the same time as your next feature release.
as a "cowboy coder" (for the most part), I'd like to pipe in: ignore everything he just said.
The only way you will ever be any good at programming is by teaching yourself, writing code in your free time, learning everything you can outside of a classroom. AND outside of work.
Too many people think that programming is something they can pick up by taking a class. I've met the results and it's not pretty.
More importantly: You do NOT need to understand the implications of any low-level operation, because as a Jr. Programmer you should not be the one making those decisions. Premature optimization leads to bad code, and too many new programmers focus on things which they are not qualified to write, or have no business writing. (I have yet to encounter a real-world situation when I have needed to write my own sorting function, hash table, or non-basic tree structure)
Always remember that for all things, someone has already written exactly what you are trying to write. Simply because it already exists, it has had more time than what you are writing has had to work out the bugs. For the love of god, don't write your own "database abstraction layer". Always assume that for any task, someone else knows how to do it better- and collaborate with other people constantly to find out how that is.
People who "aren't trained in programming", but program in their spare time are ALWAYS going to be better at writing code than people who "are trained in programming", but have never written a line of code outside of school or work.
Nope.
Why would strapping it to your face remove the necessity for head tracking? It would likely make the issue even more prominant since at that scale, eye motion is also very important. 3d on your TV, your perspective doesn't change that much when looking slightly to the left (at least, it's rather unreasonable to expect it to). Meanwhile: ever try looking to the left without moving your head? It's a neat trick that humans can do and cats can't. Your perspective completely changes, which you might be able to note by printing out something wide that looks to be in the correct perspective when looking straight ahead, then look to the side. Yes, even without moving your head, your perspective changes just as if you turned your head.
And eyes move a lot faster than heads, too. Processors in a Wii can keep up with slow head motion, but nothing that fits into a pair of glasses will keep up with your eyes darting around. They do that.
If these were simple problems to solve, VR helmets wouldn't have sucked so much.
Chrome's been out for nearly a month now and I don't see any new release any time "soon".
With such a poor release, I expected new versions to come out the same day yet here we are, weeks later, and no sign that the problems are even on Google's radar.
If I pushed a product to millions of users by linking to it from the front page of the world's most popular website, saying it was "uncrashable", and then it turned out within minutes of real-world uses that no, it's just as easy to crash as any other browser (I've yet to see a "sad tab"), or any of the other major problems, etc- I'd work towards fixing them ASAP. Where is the new release? Where is the new alpha?
Google fucked up. Forking might wake them up. All good forks get merged in the end, anyway.
Yeah, damn modern programmers always planning ahead and detecting serious flaws, brining the whole system down if needed when something unrecoverable happens. When there's a problem that means all your data is worthless, systems should just stay up and keep right on outputting those worthless numbers! Five years later, when somebody notices, you can apply a patch that makes no sense to anyone and isn't actually accurate ("better than nothing, right?")
See: http://firasd.org/weblog/2008/05/07/dilbert-average
and in processing technology to track eye motion on that scale.
Wearing a TV on your face and requiring you to stare straight ahead is NOT an improvement. With the clunkiness of the Wiimote (fun, maybe, but accurate it aint) I expect "head tracking" would be pretty much removed from the equation if you took away the [central point of focus] too.
The result: two 2-d images, on your head.