Here it is again, with the bits that don't matter (may, may be, example, preferably) removed:
"A system for facilitating online searches suggests query autocompletion strings (terms and/or phrases) to users during the query entry process, wherein the suggested strings are based on specific attributes of the particular database access system being searched. A string extraction component associated with a database access system... periodically generates a dataset that contains the autocompletion strings for the system.... The datasets are transmitted to users' computing devices,... An autocompletion client which runs on the computing devices in association with a browser uses the datasets to suggest the autocompletion strings as users enter queries that are directed to the database access system. "
Break it down: periodically we scan a database, and send (when is unspecified) the search results to the client for autocompletion.
Prior art: Any system which ran database scans to build selection lists for client autocompletion. The autocompletion would have to run on the client. This eliminates most systems, but has DEFINITELY been done before 2000. Specifically, a lot of database query apps get pre-chewed lists for autocompletion/selection. Was this done to facilitate queries? Hell yes. Look at natural language query systems. The vocabulary is sent along for pre-selection, along with DWIM functionality for error correction. Its the same thing. Other examples left to the gentle reader.
But, this is specifically to "facilitate on-line searches". So, just don't do that. "facilitate the correction of erroneous operator input" instead. Application is far broader than just "on-line searches".
"No other language has as rich a macro facility as Lisp (and its dialects)."
Not true. Take SNOBOL4. Similar to LISP, only has a single statement. And, can emulate LISP quite handily (should you so choose). Has backtracking. And, the interpreter/compiler is available allowing the language to extend itself. Indeed, I would argue that, as a macro language, SNOBOL4 is lightyears ahead of Lisp.
You are absolutely right -- I just did this in a hurry: "#define NOTHING/*=i=*/" and
#define NOTHING/*=i=*/
if (...) NOTHING; else { ...more code... }
is more like it. Now, there is a contract between the maintainer and the developer. One which I try very hard not to break. Specifically, what the code says it does is what it does. There is no need to "figure it all out" at the same time when analysing. It can be approached in a top-down fashion. A function with the name "create_mutex" is going to create a mutex, and not (say) open a file.
Beyond that, I promise that there are not tricky bits, unless specifically documented. Sure, macros have problems, but when properly used they can aid in the reading of code. The main problem macros have is they sure make it hard to write code that reads and manipulates code.
The name "ratboy" stems from a contest. We were having a patio party whose theme was "worst songs of the 70s and 80s". Each had to get up to sing. My contribution (almost the winner) was "Ben". Since then, I have been known as ratboy.
Ratboy.
PS. And I stick by my original source... and I really wish I had a friend like Ben.
I know I can change this to eliminate the empty ';', but I choose not to, because I feel that the conditional reads better as a positive statement.
[I also prefer to have the '<' or '>' in a compound conditional always point the same way:
(5 <= x) && (x <= 10)
so the conditional can be read "x between 5 and 10". I am a hacker -- thinking about code is a compulsion for me].
Now, LINT (splint which I use), complains about the empty ';' as a probable coding mistake... I redefine NOTHING as/*@i@*/; and splint no longer complains ("indent" doesn't do the right thing, so I REALLY use/*=i=*/ which indent can handle, and splint can be trained to). Plus, when I am reading the code I can tell instantly that I meant it that way. FOREVER can be for(;;) or while(1), but it is clear from the read that I mean to iterate, well, forever. Typically, I would use that as the main loop of a thread.
As to BEGIN and END -- I have coded with C (pre-ansi) on machines whose keyboards didn't have '{' and '}'. I had to construct an include file to define BEGIN and END with a binary editor. But it worked.
There may be reasons that hackers do this -- don't be immediately dismissive. [#define ONE 1 I have used as well -- in defining manifest constants for numeric types in an object oriented number stack for C++. Ok, it was a constructor: myint one = 1; but I did it to keep only one copy of the number around -- improved efficiency]. The PUSHORT style I agree is USUALLY nonsense. I can see USHORT, but the '*' notation is supported. Of course there may be an attempt to impose an object system underneath, and it may be nice to define PUSHORT as "void*" to quickly highlight all uses of the underlying implementation. So even that has a purpose.
I *am* a hacker. See, says "code monkey" on my card. And I can't READ that shit. So, here is a little snobol4 program to interpret hacker-speak. The hacker speak generator is left as an assignment for the reader...
Am I a real hacker now? Please, pretty please!?!
[I just realized, this is probably too damn obscure -- this is meant as humour].
Ratboy.
#!/usr/local/bin/snobol4 * dehack.sno * Fix up hacker speak to english &TRIM = 1 WORD = "'-" '0123456789' &UCASE &LCASE WPAT = BREAK(WORD) SPAN(WORD) . WORD :(FIRSTL) NEXTL OUTPUT = O FIRSTL O = '' LINE = INPUT:F(DONE) WORD POS(0) '1' RPOS(0):F(P2) O = O ' I':(NEXTW) P2 WORD POS(0) SPAN('012345689') RPOS(0):F(P3) O = O ' ' WORD:(NEXTW) P3 WORD POS(0) 'd00dz' RPOS(0):F(P4) O = O ' dudes':(NEXTW) P4 WORD POS(0) '1337' RPOS(0):F(P5) O = O ' elite':(NEXTW) P5 WORD POS(0) 'h4x0r3d' RPOS(0):F(P6) O = O ' hacked':(NEXTW) P6 WORD POS(0) 'h4x0r' RPOS(0):F(P7) O = O ' hacker':(NEXTW) P7 WORD = REPLACE(WORD, '307415', 'eotais') P8 WORD 'ph' = 'f':S(P8) P9 WORD 'Ph' = 'F':S(P9) O = O ' ' WORD:(NEXTW) DONE END
Normally, I wouldn't bother with AC, but this was an interesting quote:
"Other non-MS applications use the internal browser, so why should MS be forced to make it removable. Then people taht removed it would be breaking other commercial applications that use it just as another tool, like the file access APIs. You want those to be swappable too? Could you imagine if every application had to either provide their own disk access library or use "whatever the user happens to have installed" which may or many not be compatible with the original. MS owns the operating system, they have the right to make provide non-replacable APIs. There's no simpler way to state it......."
The way *I* see it is that it is my responsibility to integrate. Microsoft (and other) EULA state that I have no warantee. Personally, I *LIKE* that. If I want to fuck with the browser, I WANT TO. What rubs me the wrong way is if a vendor says: No warantee, no recourse AND LIVE WITH IT. If I can't change it, then I want some guarantee that its gonna work. And THAT'S the way to state it......
Relevant parts of Canada Copyright law, section VII (my comments are in []):
79. In this Part,
"audio recording medium"
"audio recording medium" means a recording medium, regardless of its material form, onto which a sound recording may be reproduced and that is of a kind ordinarily used by individual consumers for that purpose, excluding any prescribed kind of recording medium;
[Hard disc is not specifically mentioned, so it is ok, also consumers now "ordinarily" use hard disc for this purpose -- otherwise there wouldn't be an issue!]
[Also, copyright had to exist in Canada, or there wouldn't be an issue either, so I am going to skip that part. Note that if copyright didn't hold in Canada, it would be legal anyway!].
80. (1) Subject to subsection (2), the act of reproducing all or any substantial part of
(a) a musical work embodied in a sound recording,
(b) a performer's performance of a musical work embodied in a sound recording, or
(c) a sound recording in which a musical work, or a performer's performance of a musical work, is embodied
onto an audio recording medium for the private use of the person who makes the copy does not constitute an infringement of the copyright in the musical work, the performer's performance or the sound recording.
Limitation
(2) Subsection (1) does not apply if the act described in that subsection is done for the purpose of doing any of the following in relation to any of the things referred to in paragraphs (1)(a) to (c):
(a) selling or renting out, or by way of trade exposing or offering for sale or rental;
(b) distributing, whether or not for the purpose of trade;
(c) communicating to the public by telecommunication; or
(d) performing, or causing to be performed, in public.
[When I download a song, I don't rent, distribute, communicate to the public, or perform in the public -- I don't have to care about whether the person who PUT the songs up for download violated any of the above, I just don't violate those conditions]
In conclusion -- if there is copyright in Canada, and I have a copy of the material, I am ok as long as I use it without breaking any perscriptions. Very specifically, if I use Napster (or whatever the current du jour system is) I cannot upload material.And, that's the law (here in Canada, anyway).
"The MS protocol is at least two orders of magnitude more efficient than anything X could provide.
And mix that with the 3d hardware accelerated graphics API of longhorn, then you have graphics the way it should be."
Really... and how do you figure that?
Let's take this one step at a time:
Is the 3d going to be rendered at the application server, or in the display? If its in the display, you will have to be transferring texture maps, etc. As well, you will need more memory, and graphics capability. If transferring bitmaps, you will be where VNC is today.
Having dispensed with the 3d argument, I'll tackle the 2 orders (100x). Is the graphics language 100x denser? (hint: it isn't) It can't be compression. (hint: most people do want lossless compression for most of your screen data -- this IS possible for lossy compression, though). You did say "at least".
This would be nice. 100x improvement means that I could replace my 100BT network with Bluetooth, and see interactive reponse with 30 users doing normal engineering type stuff. Other way, I should be able to have 3000 users on my 100BT network. That WOULD be nice. All with "dumber" terminals, and a central server. Go ahead, try it. (hint: if you get this working, you will be a VERY rich person)
As to reworking the X server protocol -- ok, but give me a shim. Since it *is* networkable, I would have to upgrade SUN/Solaris and Intel/Linux (and some others) applications at the same time. Very nasty. Also, Intel/Windows (cygwin). If I had a tranlating shim then I could deploy. Its just difficult to replace the whole thing across multiple platforms and OSs at once.
If you had said "*** makes a car that get 100x the fuel economy", I would ALSO question it (that would be approximately 3000MPG for those who are still with me). Why do some folks have the need to make these wild asinine claims?
MS sees "VisiOn", and decides that that's what they need. They pre-announce, and effectively kill "VisiOn".
Apple sees Xerox PARC Smalltalk. They like it, and clone it. MS then clones Apple (except that "overlapping windows" are out). This begat Windows 1 . Does Windows borrow from X? doesn't matter -- bitblt is published as the graphics primitive for Smalltalk, Windows implements it. Adds "object windows", "events". Models MVC (which is available as the book "Smalltalk 80 - Language and Implementaton").
Good enough for you? And, I confess, I misspoke about X. But my point still stands.
As I've said before, this is nonsense. You can only build on existing work. If that work is not available in source, it is effectively lost. Hardware changes too... It isn't "innovation", it's growth. If I can't study it, I can't build on it. Between "closed software", "close hardware", and, in the US, the "DMCA" you can't effectively build on the existing body of software. What this does is force programmers into re-inventing (and hopefully avoiding patent restriction -- this makes the MS SQL patent problems delightful).
This retards software development. An open model would be much more useful. View this like a novel -- EVEN THOUGH I CAN READ AND STUDY STEVEN KING'S BOOKS, AND CAN BASE NEW FICTION ON HIS BOOKS, I JUST DON'T HAVE THE TALENT TO DISPLACE HIM. And *IF* I did have the talent, that would be to everyone's benefit and enjoyment. DEC VMS was available in source form; how many people felt inclined to "rip it off"? Unix V6 was available in source (Lyons book), again, how many people "ripped it off"? Linux was a RE-IMPLEMENTATION done for personal reasons. Turns out to be useful, and others jumped on the wagon.
Of course the industry looks very "me-too". The "free software movement" must re-implement the "closed software movement", to allow the software industry to advance. If that were not the case, monopoly lock-in would be very VERY complete (imagine - MS WORD were the ONLY word processor, and it is illegal to make another, or use its data-files for ANY purpose not sanctioned. Yes, investors in MS-WORD would be happy, but why would anyone invest ANYTHING is new features? I am a share-holder in MS, and I would SUE MS for missappropriation of share-holder equity). Besides which, MS WORD is an incremental improvement over previous Word Processors. MS had to spend more to implement, because (at least in the WP area), most WP software was proprietary (except for the best formatters - troff and TeX, but they did not address the WYSIWYG feature). Because troff and TeX are "open source", I can build on them (study how automatic hyphenation works, problems in trying to lay out pages, how to do effective typography). Knuth has even written books on the subject, which I can learn from. Where is re-usable knowledge from proprietary vendors?
What happens is that someone decides to base on the free ground knowledge, and invest money expanding this knowledge. This is then productized and sold. Of course, the product itself is "closed" in the proprietary model. The advances made cannot benefit anyone except the immediate product users. (eg. Microsoft invests a lot of money in MS WORD usability studies -- the only way I can know what the results of that research are is to compare two versions of the MS WORD product). After the product is gone, or the company is gone, the work is "lost". If the work is "locked in" no one can make use of the information (SCO's current IBM lawsuite). Does Microsoft or any other company really think that I can produce a (provably) non-infringing product based on their research? Boy, they must take my abilities in the highest esteem. Thanks guys!
Let's take your implication: that free software subculture comes out with CLONES of commercial products and not original software. How do you explain the following:
1 - The MS "Windows" windowing system is a "clone" of Smalltalk-80 and X, both of which are free in the sense that source was available.
2 - The software industry uses Berkely sockets. Which of course is "free". All common protocols are based on this.
3 - SQL is a standard. Yes, an "open implementation" wasn't available, but the spec is open.
4 - The kernel design of Windows NT is based on "open" systems. VMS source was available for study, and was Unix.
I can't think of anything lasting coming from closed source initiatives. With an exception -- the spreadsheet.
Do you have any other examples?
We base our progress on the work of others (history). If that history is locked up, we are doomed to re-inventing it. Which, in the long run, doesn't give us as much progress.
I have been programming (managing, designing) for a living since 1976. I have hardware patents for graphics chips (and, yes, the chips where actually made). I have designed, built and sold disc controllers, SCSI (SASI) interfaces, and some other stuff. I know typesetting and word processing. I have produced software and hardware that has been named "best of breed" by Seybold.
In other words, I am a very experienced developer. I am not a business person; I have made millions and LOST them again over my career.
Now, where is this going? I put my resume in to Microsoft (back in '99), and received a phone interview. In the interview, I was asked two questions -- and one was: "If you take apart a clock to fix it, put it back together, and have some screws left over, what do you do"? I asked -- does the clock work? Answer: yes. I asked -- is it my clock? Answer: yes. My answer: put the extra screws in a bag and tape it to the back of the clock. Put the clock into service.
I was not asked about ANYTHING on my resume, which I found interesting. The other question was "Why do you want to work for Microsoft"? My answer: I want to be with an organization that has potential and is aggressive in producing results.
I was not invited for an interview.
I have wondered, on and off, what was the desired answer to the question?
The problem is file system meta-data. Competition is with journaling fs's! Meta-data must be updated, and if it is to disc, the operation is slow. It is not clear to me at all that RAM caching as you describe is an answer. At some point the data must be put to permanent storage. And this is where the bottleneck comes in. Unless you want all your work to be eaten on a power failure.
You're a small band. Someone, somewhere, rips your stuff and puts it on-line. I download it in Canada. I put it onto a CD-R. You know what? You are screwed. You CAN'T prosecute me (well, if I entered your country, you could). And if you happen to be Canadian as well? Too f*cking bad. Its your problem. So, get active, and protest. Actually, US copyright section 100x (the 1992 law) looks interesting. I just have to make sure that generational (SCM) isn't being violated, and that the royalty has been paid. After this, apparently you CAN'T be prosecuted for copying anymore. A flat (small) percentage on a hard drive that you intend to use for MP3 storage. Remit the cheque, and go nuts! Since there is NO SCM on CDs (by design), it isn't being circumvented. Any US Copyright lawyers want to comment?
Of course paying the levy in Canada gives you the right to use copyrighted works. The law changed in 1999 -- personal copying is permitted. It perfectly legal in Canada to DOWNLOAD music. Can't UPLOAD it though... (remember "personal" copying). Still, lots of people upload, so we are PERFECTLY legal. Must burn music onto CD, or tape, it's not legal to leave it on the hard drive...
So, the private key never leaves the chip? And... how did it actually get there? Does each chip make its own keys? How do we 'vette that operation? Or is the key seeded by the hardware manufacturer. And could the manufacturer be bought? Let me be blunt -- I WANT CONTROL OF MY KEYS. *I* don't trust the chip... and yes, I work at a company that designs microchips, that have security in them. And, if I don't trust the chips, then I can't trust the software. That may change from another perspective. That is RIAA et al. may trust my computer -- but I DON'T. If I give RIAA an encryption key, and they encrypt content to that key, I will trust it. If the RIAA gives me a public key, and says "here is content from me, decrypt with that key and it is proven" -- it isn't proven. All you have is the statement that they are willing to trust the crypto system *AND* they are willing to swear that it is in their interest to avow that the private key has not been comprimised.
Now, if I DON'T KNOW or HAVE ACCESS TO the private key, *I* cannot avow that the key has not been comprimised. It might have been. The content supplier actually knows more about this that I do. They may trust, but I don't.
Why do I have separate close calls for sockets under Windows? And what about select() -- can't seem to pass in NON-sockets. I haven't tried passing in a socket to (say) WaitSingleObject(), but the types are wrong, so it pro'lly won't even compile right. And this is CONSISTENT? Foo on you! Sockets don't even work as an IPC mechanism, which is a real pain... and there isn't even a socketpair() function.
Now, Linux (default) doesn't do message queues, and the pthread mutex were process specific, and didn't work 'twixt process (I should be able to mmap() in memory, and use the same mutex in two different processes... with the addition of a parameter!) Natch' Solaris supports ALL of that.
So, my OS ranking:
1 - Solaris 2 - Linux 3 - Windows
Interesting, because my code supports stuff in the EXACT opposite order. And, I have portability kludges all over the place. Piss on this!
Take Windows and shove it. Then, merge Linux and Solaris. Yeah... that's the plan.
Fascinating -- the ONLY thing I get out of contributing to Open Source Code is -- developer's ego.Thanks for labelling me "stupid". Just for that, you can't use my open-source software.
Do you want to put some money where your mouth is? Pay to have the problems fixed.
Also, what is "mostly works"?
I have a story to tell. It's a true story. The "user" may have been YOU (the attitude is right).
Years ago, I released an 8087 emulator (emu87.zip), that allowed the use of Turbo Pascal 8087 and AutoCad on machines which did not have the 8087 FPU chip. I did it because my software was built with TP 8087, and one of the marketing droids has purchased a laptop for demos that DIDN'T allow the chip to be added. I released the software -- and, as far as I could tell -- it built a user-base of several thousand users. I did not release source for the product, because it used the '87 compatible fpu library from a C compiler. The rest of the code was 50 or so lines (just some glue).
I asked for no money for this software. After all, it wasn't really worth anything...
Five (5) years later, some yahoo contacts me with a threat! Yes, a customer of his had aquired a NEW computer, and this software would no longer work! If I didn't IMMEDIATELY fix the problem (not mentioned was what the problem actually was), I would be sued! Or worse...
Hey, if he had sent me a "thank you" note in the first place, I might have responded positively. As it was, I told him to take a flying f*ck at a rolling dougnut. Remember, the ONLY thing I could EVER get out of the release of this utility was ego.
Timed myself doing it -- took 4-5 seconds when I was paying attention, 10 seconds when I wasn't.
The sequence MAY be smaller, depends on OTHER file names -- feedback is a [beep] if it isn't unique yet.
Ratboy.
PS. Yes, Unix systems were available (SCO, IDRIS, IBM, XENIS, others) The only reason that they were not used is that the client insisted on Macs. And, this was in 1989/90.
Yes, I *have* used the "Classic Mac OS Finder". And quite detested it. I tried to use it in a production environment, where the clients demanded use of Mac technology. Problem was... files (work) is scattered over a network of machines, with automatically generated names (001-00156, 001-00157, etc. -- names chosen to be useful within the problem domain). Now, we have 300-400 such files per unit, and several thousand units on the go... There is a problem with a file, now, drill down to find and fix (or select based on templates). Of course, I attempted this BEFORE scripting was available on Macs. Being an old Unix guy, I just ASSUMED that this type of operation was tivial--after all this is a COMPUTER, right?
As a result of this experience, I NEVER USED A MAC AGAIN. Maybe now, with OSX I'll think about it.
I don't have any problems with a "spatial" interface, for those users that can't deal with anything else. Go ahead, put it in... In fact, I use a "spatial" interface on at least one of my machines -- I have folders on the desktop that refer to directories on the 'net. So, I don't have to think about opening up my status folder to update work status. You know what? It probably does save me time. But for the "real work", I still go back to a symbolic command interface.
Slashdot Mirror
Here it is again, with the bits that don't matter (may, may be, example, preferably) removed:
... periodically generates a dataset that contains the autocompletion strings for the system. ... The datasets are transmitted to users' computing devices, ... An autocompletion client which runs on the computing devices in association with a browser uses the datasets to suggest the autocompletion strings as users enter queries that are directed to the database access system. "
"A system for facilitating online searches suggests query autocompletion strings (terms and/or phrases) to users during the query entry process, wherein the suggested strings are based on specific attributes of the particular database access system being searched. A string extraction component associated with a database access system
Break it down: periodically we scan a database, and send (when is unspecified) the search results to the client for autocompletion.
Prior art: Any system which ran database scans to build selection lists for client autocompletion. The autocompletion would have to run on the client. This eliminates most systems, but has DEFINITELY been done before 2000. Specifically, a lot of database query apps get pre-chewed lists for autocompletion/selection. Was this done to facilitate queries? Hell yes. Look at natural language query systems. The vocabulary is sent along for pre-selection, along with DWIM functionality for error correction. Its the same thing. Other examples left to the gentle reader.
But, this is specifically to "facilitate on-line searches". So, just don't do that. "facilitate the correction of erroneous operator input" instead. Application is far broader than just "on-line searches".
Someone want to patent that?
Ratboy.
"No other language has as rich a macro facility as Lisp (and its dialects)."
Not true. Take SNOBOL4. Similar to LISP, only has
a single statement. And, can emulate LISP quite handily (should you so choose). Has backtracking. And, the interpreter/compiler is available allowing the language to extend itself. Indeed, I would argue that, as a macro language, SNOBOL4 is lightyears ahead of Lisp.
Ratboy.
You are absolutely right -- I just did this in a hurry: "#define NOTHING /*=i=*/" and
/*=i=*/
...more code...
#define NOTHING
if (...)
NOTHING;
else {
}
is more like it. Now, there is a contract between the maintainer and the developer. One which I try very hard not to break. Specifically, what the code says it does is what it does. There is no need to "figure it all out" at the same time when analysing. It can be approached in a top-down fashion. A function with the name "create_mutex" is going to create a mutex, and not (say) open a file.
Beyond that, I promise that there are not tricky bits, unless specifically documented. Sure, macros have problems, but when properly used they can aid in the reading of code. The main problem macros have is they sure make it hard to write code that reads and manipulates code.
Ratboy
The name "ratboy" stems from a contest. We were having a patio party whose theme was "worst songs of the 70s and 80s". Each had to get up to sing. My contribution (almost the winner) was "Ben". Since then, I have been known as ratboy.
Ratboy.
PS. And I stick by my original source... and I really wish I had a friend like Ben.
Redefining the language to better meet the problem is one of the hallmarks of good (artistic) programming. And it has other purposes. To give you one:
;
...
/*@i@*/; and splint no longer complains ("indent" doesn't do the right thing, so I REALLY use /*=i=*/ which indent can handle, and splint can be trained to). Plus, when I am reading the code I can tell instantly that I meant it that way. FOREVER can be for(;;) or while(1), but it is clear from the read that I mean to iterate, well, forever. Typically, I would use that as the main loop of a thread.
#define NOTHING
...
if ((flag & VAL1) || (flag & VAL2)
NOTHING;
else
I know I can change this to eliminate the empty
';', but I choose not to, because I feel that the conditional reads better as a positive statement.
[I also prefer to have the '<' or '>' in a compound conditional always point the same way:
(5 <= x) && (x <= 10)
so the conditional can be read "x between 5 and 10". I am a hacker -- thinking about code is a compulsion for me].
Now, LINT (splint which I use), complains about the empty ';' as a probable coding mistake... I redefine NOTHING as
As to BEGIN and END -- I have coded with C (pre-ansi) on machines whose keyboards didn't have '{' and '}'. I had to construct an include file to define BEGIN and END with a binary editor. But it worked.
There may be reasons that hackers do this -- don't be immediately dismissive. [#define ONE 1 I have used as well -- in defining manifest constants for numeric types in an object oriented number stack for C++. Ok, it was a constructor: myint one = 1; but I did it to keep only one copy of the number around -- improved efficiency]. The PUSHORT style I agree is USUALLY nonsense. I can see USHORT, but the '*' notation is supported. Of course there may be an attempt to impose an object system underneath, and it may be nice to define PUSHORT as "void*" to quickly highlight all uses of the underlying implementation. So even that has a purpose.
Ratboy.
I *am* a hacker. See, says "code monkey" on my card. And I can't READ that shit. So, here is a little snobol4 program to interpret hacker-speak. The hacker speak generator is left as an assignment for the reader...
/usr/local/bin/snobol4
:(FIRSTL) :F(DONE) :F(P2) :(NEXTW) :F(P3) :(NEXTW) :F(P4) :(NEXTW) :F(P5) :(NEXTW) :F(P6) :(NEXTW) :F(P7) :(NEXTW) :S(P8) :S(P9) :(NEXTW)
Am I a real hacker now? Please, pretty please!?!
[I just realized, this is probably too damn obscure -- this is meant as humour].
Ratboy.
#!
* dehack.sno
* Fix up hacker speak to english
&TRIM = 1
WORD = "'-" '0123456789' &UCASE &LCASE
WPAT = BREAK(WORD) SPAN(WORD) . WORD
NEXTL OUTPUT = O
FIRSTL O = ''
LINE = INPUT
WORD POS(0) '1' RPOS(0)
O = O ' I'
P2 WORD POS(0) SPAN('012345689') RPOS(0)
O = O ' ' WORD
P3 WORD POS(0) 'd00dz' RPOS(0)
O = O ' dudes'
P4 WORD POS(0) '1337' RPOS(0)
O = O ' elite'
P5 WORD POS(0) 'h4x0r3d' RPOS(0)
O = O ' hacked'
P6 WORD POS(0) 'h4x0r' RPOS(0)
O = O ' hacker'
P7 WORD = REPLACE(WORD, '307415', 'eotais')
P8 WORD 'ph' = 'f'
P9 WORD 'Ph' = 'F'
O = O ' ' WORD
DONE
END
Normally, I wouldn't bother with AC, but this was an interesting quote:
"Other non-MS applications use the internal browser, so why should MS be forced to make it removable. Then people taht removed it would be breaking other commercial applications that use it just as another tool, like the file access APIs. You want those to be swappable too? Could you imagine if every application had to either provide their own disk access library or use "whatever the user happens to have installed" which may or many not be compatible with the original. MS owns the operating system, they have the right to make provide non-replacable APIs. There's no simpler way to state it......."
The way *I* see it is that it is my responsibility to integrate. Microsoft (and other) EULA state that I have no warantee. Personally, I *LIKE* that. If I want to fuck with the browser, I WANT TO. What rubs me the wrong way is if a vendor says: No warantee, no recourse AND LIVE WITH IT. If I can't change it, then I want some guarantee that its gonna work. And THAT'S the way to state it......
Ratboy
No, not illegal:
d -3 8379
Relevant parts of Canada Copyright law, section VII (my comments are in []):
79. In this Part,
"audio recording medium"
"audio recording medium" means a recording medium, regardless of its material form, onto which a sound recording may be reproduced and that is of a kind ordinarily used by individual consumers for that purpose, excluding any prescribed kind of recording medium;
[Hard disc is not specifically mentioned, so it is ok, also consumers now "ordinarily" use hard disc for this purpose -- otherwise there wouldn't be an issue!]
[Also, copyright had to exist in Canada, or there wouldn't be an issue either, so I am going to skip that part. Note that if copyright didn't hold in Canada, it would be legal anyway!].
80. (1) Subject to subsection (2), the act of reproducing all or any substantial part of
(a) a musical work embodied in a sound recording,
(b) a performer's performance of a musical work embodied in a sound recording, or
(c) a sound recording in which a musical work, or a performer's performance of a musical work, is embodied
onto an audio recording medium for the private use of the person who makes the copy does not constitute an infringement of the copyright in the musical work, the performer's performance or the sound recording.
Limitation
(2) Subsection (1) does not apply if the act described in that subsection is done for the purpose of doing any of the following in relation to any of the things referred to in paragraphs (1)(a) to (c):
(a) selling or renting out, or by way of trade exposing or offering for sale or rental;
(b) distributing, whether or not for the purpose of trade;
(c) communicating to the public by telecommunication; or
(d) performing, or causing to be performed, in public.
[When I download a song, I don't rent, distribute, communicate to the public, or perform in the public -- I don't have to care about whether the person who PUT the songs up for download violated any of the above, I just don't violate those conditions]
In conclusion -- if there is copyright in Canada, and I have a copy of the material, I am ok as long as I use it without breaking any perscriptions. Very specifically, if I use Napster (or whatever the current du jour system is) I cannot upload material.And, that's the law (here in Canada, anyway).
http://laws.justice.gc.ca/en/C-42/38266.html#ri
Ratboy.
"The MS protocol is at least two orders of magnitude more efficient than anything X could provide.
And mix that with the 3d hardware accelerated graphics API of longhorn, then you have graphics the way it should be."
Really... and how do you figure that?
Let's take this one step at a time:
Is the 3d going to be rendered at the application server, or in the display? If its in the display, you will have to be transferring texture maps, etc. As well, you will need more memory, and graphics capability. If transferring bitmaps, you will be where VNC is today.
Having dispensed with the 3d argument, I'll tackle the 2 orders (100x). Is the graphics language 100x denser? (hint: it isn't) It can't be compression. (hint: most people do want lossless compression for most of your screen data -- this IS possible for lossy compression, though). You did say "at least".
This would be nice. 100x improvement means that I could replace my 100BT network with Bluetooth, and see interactive reponse with 30 users doing normal engineering type stuff. Other way, I should be able to have 3000 users on my 100BT network. That WOULD be nice. All with "dumber" terminals, and a central server. Go ahead, try it. (hint: if you get this working, you will be a VERY rich person)
As to reworking the X server protocol -- ok, but give me a shim. Since it *is* networkable, I would have to upgrade SUN/Solaris and Intel/Linux (and some others) applications at the same time. Very nasty. Also, Intel/Windows (cygwin). If I had a tranlating shim then I could deploy. Its just difficult to replace the whole thing across multiple platforms and OSs at once.
If you had said "*** makes a car that get 100x the fuel economy", I would ALSO question it (that would be approximately 3000MPG for those who are still with me). Why do some folks have the need to make these wild asinine claims?
Ratboy666
The scariest and best part is when the INTERVIEWER actually confused the RIAA with the Government. That sure gives the RIAA some big clankin' balls.
Next... RIAA orders bombing of Canada, because its acutally legal to download music here!
Ratboy666.
Actually, I will trace the history as follows:
MS sees "VisiOn", and decides that that's what they need. They pre-announce, and effectively kill "VisiOn".
Apple sees Xerox PARC Smalltalk. They like it, and clone it. MS then clones Apple (except that "overlapping windows" are out). This begat Windows 1 . Does Windows borrow from X? doesn't matter -- bitblt is published as the graphics primitive for Smalltalk, Windows implements it. Adds "object windows", "events". Models MVC (which is available as the book "Smalltalk 80 - Language and Implementaton").
Good enough for you? And, I confess, I misspoke about X. But my point still stands.
Ratboy
Begin rant...
As I've said before, this is nonsense. You can only build on existing work. If that work is not available in source, it is effectively lost. Hardware changes too... It isn't "innovation", it's growth. If I can't study it, I can't build on it. Between "closed software", "close hardware", and, in the US, the "DMCA" you can't effectively build on the existing body of software. What this does is force programmers into re-inventing (and hopefully avoiding patent restriction -- this makes the MS SQL patent problems delightful).
This retards software development. An open model would be much more useful. View this like a novel -- EVEN THOUGH I CAN READ AND STUDY STEVEN KING'S BOOKS, AND CAN BASE NEW FICTION ON HIS BOOKS, I JUST DON'T HAVE THE TALENT TO DISPLACE HIM. And *IF* I did have the talent, that would be to everyone's benefit and enjoyment. DEC VMS was available in source form; how many people felt inclined to "rip it off"? Unix V6 was available in source (Lyons book), again, how many people "ripped it off"? Linux was a RE-IMPLEMENTATION done for personal reasons. Turns out to be useful, and others jumped on the wagon.
Of course the industry looks very "me-too". The "free software movement" must re-implement the "closed software movement", to allow the software industry to advance. If that were not the case, monopoly lock-in would be very VERY complete (imagine - MS WORD were the ONLY word processor, and it is illegal to make another, or use its data-files for ANY purpose not sanctioned. Yes, investors in MS-WORD would be happy, but why would anyone invest ANYTHING is new features? I am a share-holder in MS, and I would SUE MS for missappropriation of share-holder equity). Besides which, MS WORD is an incremental improvement over previous Word Processors. MS had to spend more to implement, because (at least in the WP area), most WP software was proprietary (except for the best formatters - troff and TeX, but they did not address the WYSIWYG feature). Because troff and TeX are "open source", I can build on them (study how automatic hyphenation works, problems in trying to lay out pages, how to do effective typography). Knuth has even written books on the subject, which I can learn from. Where is re-usable knowledge from proprietary vendors?
What happens is that someone decides to base on the free ground knowledge, and invest money expanding this knowledge. This is then productized and sold. Of course, the product itself is "closed" in the proprietary model. The advances made cannot benefit anyone except the immediate product users. (eg. Microsoft invests a lot of money in MS WORD usability studies -- the only way I can know what the results of that research are is to compare two versions of the MS WORD product). After the product is gone, or the company is gone, the work is "lost". If the work is "locked in" no one can make use of the information (SCO's current IBM lawsuite). Does Microsoft or any other company really think that I can produce a (provably) non-infringing product based on their research? Boy, they must take my abilities in the highest esteem. Thanks guys!
Enough of a rant.
Ratboy
Let's take your implication: that free software subculture comes out with CLONES of commercial products and not original software. How do you explain the following:
1 - The MS "Windows" windowing system is a "clone" of Smalltalk-80 and X, both of which are free in the sense that source was available.
2 - The software industry uses Berkely sockets. Which of course is "free". All common protocols are based on this.
3 - SQL is a standard. Yes, an "open implementation" wasn't available, but the spec is open.
4 - The kernel design of Windows NT is based on "open" systems. VMS source was available for study, and was Unix.
I can't think of anything lasting coming from closed source initiatives. With an exception -- the spreadsheet.
Do you have any other examples?
We base our progress on the work of others (history). If that history is locked up, we are doomed to re-inventing it. Which, in the long run, doesn't give us as much progress.
Ratboy.
"a threat to SCO in the heavy duty 64-bit space."
Um... how is that? Does SCO have a 64 bit product?
Ratboy.
I have been programming (managing, designing) for a living since 1976. I have hardware patents for graphics chips (and, yes, the chips where actually made). I have designed, built and sold disc controllers, SCSI (SASI) interfaces, and some other stuff. I know typesetting and word processing. I have produced software and hardware that has been named "best of breed" by Seybold.
In other words, I am a very experienced developer. I am not a business person; I have made millions and LOST them again over my career.
Now, where is this going? I put my resume in to Microsoft (back in '99), and received a phone interview. In the interview, I was asked two questions -- and one was: "If you take apart a clock to fix it, put it back together, and have some screws left over, what do you do"? I asked -- does the clock work? Answer: yes. I asked -- is it my clock? Answer: yes. My answer: put the extra screws in a bag and tape it to the back of the clock. Put the clock into service.
I was not asked about ANYTHING on my resume, which I found interesting. The other question was "Why do you want to work for Microsoft"? My answer: I want to be with an organization that has potential and is aggressive in producing results.
I was not invited for an interview.
I have wondered, on and off, what was the desired answer to the question?
Ratboy
The problem is file system meta-data. Competition is with journaling fs's! Meta-data must be updated, and if it is to disc, the operation is slow. It is not clear to me at all that RAM caching as you describe is an answer. At some point the data must be put to permanent storage. And this is where the bottleneck comes in. Unless
you want all your work to be eaten on a power failure.
You're a small band. Someone, somewhere, rips your stuff and puts it on-line. I download it in Canada. I put it onto a CD-R. You know what? You are screwed. You CAN'T prosecute me (well, if I entered your country, you could). And if you happen to be Canadian as well? Too f*cking bad. Its your problem. So, get active, and protest. Actually, US copyright section 100x (the 1992 law) looks interesting. I just have to make sure that generational (SCM) isn't being violated, and that the royalty has been paid. After this, apparently you CAN'T be prosecuted for copying anymore. A flat (small) percentage on a hard drive that you intend to use for MP3 storage. Remit the cheque, and go nuts! Since there is NO SCM on CDs (by design), it isn't being circumvented. Any US Copyright lawyers want to comment?
Ratboy.
Of course paying the levy in Canada gives you the right to use copyrighted works. The law changed in 1999 -- personal copying is permitted. It perfectly legal in Canada to DOWNLOAD music. Can't UPLOAD it though... (remember "personal" copying). Still, lots of people upload, so we are PERFECTLY legal. Must burn music onto CD, or tape, it's not legal to leave it on the hard drive...
Ratboy.
So, the private key never leaves the chip? And... how did it actually get there? Does each chip make its own keys? How do we 'vette that operation? Or is the key seeded by the hardware manufacturer. And could the manufacturer be bought? Let me be blunt -- I WANT CONTROL OF MY KEYS. *I* don't trust the chip... and yes, I work at a company that designs microchips, that have security in them. And, if I don't trust the chips, then I can't trust the software. That may change from another perspective. That is RIAA et al. may trust my computer -- but I DON'T. If I give RIAA an encryption key, and they encrypt content to that key, I will trust it. If the RIAA gives me a public key, and says "here is content from me, decrypt with that key and it is proven" -- it isn't proven. All you have is the statement that they are willing to trust the crypto system *AND* they are willing to swear that it is in their interest to avow that the private key has not been comprimised.
Now, if I DON'T KNOW or HAVE ACCESS TO the private key, *I* cannot avow that the key has not been comprimised. It might have been. The content supplier actually knows more about this that I do. They may trust, but I don't.
Ratboy.
Here's a troll...
Why do I have separate close calls for sockets under Windows? And what about select() -- can't seem to pass in NON-sockets. I haven't tried passing in a socket to (say) WaitSingleObject(), but the types are wrong, so it pro'lly won't even compile right. And this is CONSISTENT? Foo on you! Sockets don't even work as an IPC mechanism, which is a real pain... and there isn't even a socketpair() function.
Now, Linux (default) doesn't do message queues, and the pthread mutex were process specific, and didn't work 'twixt process (I should be able to mmap() in memory, and use the same mutex in two different processes... with the addition of a parameter!) Natch' Solaris supports ALL of that.
So, my OS ranking:
1 - Solaris
2 - Linux
3 - Windows
Interesting, because my code supports stuff in the EXACT opposite order. And, I have portability kludges all over the place. Piss on this!
Take Windows and shove it. Then, merge Linux and Solaris. Yeah... that's the plan.
Ratboy666
To keep the BSA turkeys happy, you need:
1 - License card and
2 - A receipt
If Microsoft "gives" you software for free, please remember to ask for the license and a receipt -- EVEN IF ITS FREE.
Ratboy.
Fascinating -- the ONLY thing I get out of contributing to Open Source Code is -- developer's ego.Thanks for labelling me "stupid". Just for that, you can't use my open-source software.
Do you want to put some money where your mouth is? Pay to have the problems fixed.
Also, what is "mostly works"?
I have a story to tell. It's a true story. The "user" may have been YOU (the attitude is right).
Years ago, I released an 8087 emulator (emu87.zip), that allowed the use of Turbo Pascal 8087 and AutoCad on machines which did not have the 8087 FPU chip. I did it because my software was built with TP 8087, and one of the marketing droids has purchased a laptop for demos that DIDN'T allow the chip to be added. I released the software -- and, as far as I could tell -- it built a user-base of several thousand users. I did not release source for the product, because it used the '87 compatible fpu library from a C compiler. The rest of the code was 50 or so lines (just some glue).
I asked for no money for this software. After all, it wasn't really worth anything...
Five (5) years later, some yahoo contacts me with a threat! Yes, a customer of his had aquired a NEW computer, and this software would no longer work! If I didn't IMMEDIATELY fix the problem (not mentioned was what the problem actually was), I would be sued! Or worse...
Hey, if he had sent me a "thank you" note in the first place, I might have responded positively. As it was, I told him to take a flying f*ck at a rolling dougnut. Remember, the ONLY thing I could EVER get out of the release of this utility was ego.
Ratboy.
Sorry to be replying to myself: Slashdot ate part of my message: Here is the sequence (tab) and (return) refer to the keypress:
mv Pi(tab)Mo(tab)Ma(tab)..(return)
mv De(tab)Sm(tab)Be(tab)/mu(tab)(return)
Ratboy
Ok, I'll take your challenge:
mv PiMoMa..
mv DeSmBe/mu
should do it. 32 keystrokes, no mouse.
Timed myself doing it -- took 4-5 seconds when I was paying attention, 10 seconds when I wasn't.
The sequence MAY be smaller, depends on OTHER file names -- feedback is a [beep] if it isn't unique yet.
Ratboy.
PS. Yes, Unix systems were available (SCO, IDRIS, IBM, XENIS, others) The only reason that they were not used is that the client insisted on Macs. And, this was in 1989/90.
Ratboy
Yes, I *have* used the "Classic Mac OS Finder". And quite detested it. I tried to use it in a production environment, where the clients demanded use of Mac technology. Problem was... files (work) is scattered over a network of machines, with automatically generated names (001-00156, 001-00157, etc. -- names chosen to be useful within the problem domain). Now, we have 300-400 such files per unit, and several thousand units on the go... There is a problem with a file, now, drill down to find and fix (or select based on templates). Of course, I attempted this BEFORE scripting was available on Macs. Being an old Unix guy, I just ASSUMED that this type of operation was tivial--after all this is a COMPUTER, right?
As a result of this experience, I NEVER USED A MAC AGAIN. Maybe now, with OSX I'll think about it.
I don't have any problems with a "spatial" interface, for those users that can't deal with anything else. Go ahead, put it in... In fact, I use a "spatial" interface on at least one of my machines -- I have folders on the desktop that refer to directories on the 'net. So, I don't have to think about opening up my status folder to update work status. You know what? It probably does save me time. But for the "real work", I still go back to a symbolic command interface.
Ratboy.