>>4
Open standards are great, but a lot harder to do than open source software.
I'm really pissed about how the über creator of most shameful ActiveX bugs doesn't want WebGL, accusing it has lots of security flaws, just because they already have their already proven proprietary 3D browser API.
Name:
Anonymous2011-11-17 21:50
(written-in English (all standards))
(is english (crappy-ambiguous natural-language))
(should-be-replaced-with LISP English)
(at-least ((everywhere force) SEXPs))
>>12 Is There An Alternative to “Considered Harmful” Essays?
I propose the single best alternative: "If it ain't Lisp, it's crap." eassays, that will, in plain and simple words, explain to the reader, why he is a shiteating cocksucker blockhead.
>>13
"If it ain't Lisp, it's crap" essays are generally very compact, much like the Lisp language itself. As they build of of the unrivaled truths of the language, they need not diverge into complexity.
Take for example the following essay, by Andrei Alexandrescu, titled ``The Language I Wish I Had Invented.''
There's only one answer, really. Lisp. I mean I'm not even kidding. Lisp is the one language that has said the first and last word on so many fields, it's pretty much amazing. It has introduced garbage collection when it was almost impractical to do it. It has introduced functional programming, it has introduced a lot of notions that people have ever since rediscovered many times over. So, I wish I had invested lisp.
"Dealing with complexity is an inefficient and unnecessary waste of time, attention and mental energy. There is never any justification for things being complex when they could be simple."
Dr. Edward de Bono
Name:
Anonymous2011-11-18 3:47
Considered hamful.
Name:
Anonymous2011-11-18 4:24
OPEN ASSHOLES, NOT OPEN STANDARDS
Name:
Anonymous2011-11-18 8:13
Lisp is shit.
Name:
Anonymous2011-11-18 9:14
BSD is the best kind of software standard, they tap into cost/benefit factors.
Standards are broken by design, they're only followed for a few reasons:
1. marketing ("this toaster works like your previous toaster!")
2. government ("this toaster can plug into your wall!")
3. safety ("ul and csa checked, this toaster won't kill you!")
Since software doesn't catch fire and governments haven't seriously regulated anything software related. Software standards are just marketing, so they are super-broken:
1. De jure standards for marketing purposes are worthless in a pure market based environment, businesses continuously differentiate to stay alive. Every de facto standard happens to be because some previous big cat made something everyone bought or pirated and the next big cat want its users.
2. The biggest cat will subvert the standards in every way possible because they can and they profit.
Name:
Anonymous2011-11-18 9:18
>>20 BSD is the best kind of software standard, they tap into cost/benefit factors.
Agreed. As much as I hate Theo, I find his philosophy very agreeable. He's happy for companies to do whatever the fuck they want with OpenSSH because it's better than them writing their own broken, proprietary implementations.
The best standards are ones that are accepted because they're high quality, not just because of widespread use or because Standards Body X says to use them.
ISO9660 mentions binary zeroes. That's the kind of things you find in standards.
Name:
Anonymous2011-11-18 22:18
>>22
That's inherent to the marketing aspect of standards though, which is why getting to use that USB logo on a product's packaging costs a few grand.
Name:
Anonymous2011-11-19 1:22
>>23
Binary zeroes means all-bits-zero ((unsigned char)0) as opposed to zero characters ('0'), or integers/floating-point numbers that compare equal to zero, but are not all-bits-zero (such as -0.0 or an integer with non-zero padding bits). In a standard, this is important because developers could interpret unqualified "zero" as 0x30, 0xF0, or 0x80000000, and not be incorrect.
Name:
Anonymous2011-11-19 2:57
>>25
Retarded logic. Do they know about context or common sense?
The thing is is that there are many contexts for the number zero, '0', +0.0, -0.0, and then there are also different representations of zero based upon what the cpu is using for integer arithmetic. It could be unsigned, or signed using 2's compliment, in which case a zero value would be binary zero. Or it could be signed magnitude, where one bit is reserved to represent sign and the rest of the bits represent the absolute value of the number. In this case, there is both a negative and a positive version of zero, both having distinct representations in binary. It is good to establish which zero you are talking about, like a block of all zeroed bits, to avoid ambiguity.
>>25-30
The context of those ``binary zeroes'' consists of middle-endian unsigned numbers. ISO9660 doesn't even specify a character set for the fields in which they're used, so there's no risk of using '0'.
That's not about standards, it's about standards bodies and companies. There are standards bodies that make you pay through the nose to use their logo, then there are standards bodies like the IETF that just make standards so that stuff interoperates and don't charge anything for it.
Name:
332011-11-19 15:28
Those are pretty different things though. The IETF is sort of a mailing list where stuff happens and text files come out of it so the next guy can read it to make their own ircd... and a pretty large number of popular protocols don't have RFCs. I'd say mostly programmers who do sockets would know what the IETF and its RFCs are so it doesn't really have any actual marketing weight logo-wise. OTOH, someone who downloads something that advertises itself to be an IRC client and it doesn't actually connect to anything IRC, the soft will just be collectively deleted, downvotes will be given and bad reviews will surface.
USB forum (and the rest of these types of things) defines the standard through consensus between the paid members I'd think, but also provides a bunch of device class specific utilities, test suites, a registry to prevent VID collisions, etc. Consumers look for the logo and they'll know it will plug in and probably work (not as much in a lot of cases now as it was when it was new I'd assume, since it is absolutely ubiquitous, except maybe for 3.0 and the other specialist ones).
Cleary >>34 doesn't know anything outside of SICP.
Name:
Anonymous2011-11-20 0:30
>>36
Those are transport layer of which they are all obviously covered. Application layer is a different story:
▶ Obviously every online multiplayer game out there, but we'll say that these don't count.
▶ Most (all?) P2P protocols that I'm aware of, I'm surprised myself here, a search for `bittorrent rfc' brings up a student project (to create a RFC?), the Wikipedia page just because its Wikipedia, a forum post of a dude complaining about the lack of a bittorrent RFC, and then whatever.
▶ Every IM protocol other than XMPP, that I'm aware of.
▶ All sorts of random shit like Flash player's use of TCP port 843.
There's likely more but thats what was off the top of my head at this point.
I'm not sure what your point is. I was simply refuting the statement that "standards are broken by design" argument of >>20 by pointing out IETF as an example of a standards body that has nothing to do with "marketing", "government" or "safety".
In IETF a bunch of people (everyone at IETF represents only themselves, not their organizations) get together to agree on a common standard so that stuff interoperates. Wether there's an RFC for this or that protocol is irrelevant.