I missed this the first time round and probably should have posed this as a question, but its a bit late now.
Basically, what does Reason offer over running gcc -Wall -pedantic -anything else, running lint on the code, and then running the code with something like valgrind to detect dynamic memory problems? Okay, its a single tool as opposed to many of these and probably produces nicer reports, but what I think I'm really asking is, for the poor student like me, is there anything similar/have I missed out any really useful tools from the above (lint/valgrind/gcc with all warning options)?
So is the fork going to be called...
on
XFree86 Politics
·
· Score: 1
OpenXFree86, and concentrate on proactive security whilst still being easy to use?
see 'shell company' and start formulating a reply along the lines of.. 'yeah, theres plenty, but few will let you run eggdrops/bnc servers.. etc..etc..'?
Wow.. that sounds amazing, especially considering its computer generated. So now I have to tell everyone that a computer can not only beat me at board games, but it can improvise better than I can as well:(
Admittedly, jazz is a lot more random than other styles, and the chords for some (all?) of the songs were predetermined, but still, its the first computer generated music I've heard that didnt sound 'weird', unlike the ones from the toplevel article here, parts of which sound like they would make really good intros/backings to a tune, but needs something to make the music more melodic (I guess that where the GA comes in).
I'd be interested to know how many generations it takes for most of the tunes, and what fitness function he used (or if it was just.. listen to several and pick the best few).. guess its reading time:)
I'd argue that whether it is viral isnt (or needn't be) the issue here.
Both word documents and the GPL license prevent you from making use of {documents|code} in certain circumstances. (Yes I know the GPL grants you extra rights on copyright, but all licenses have conditions, thereby preventing use).
But the important thing is - YOU DO NOT HAVE TO USE EITHER in your work.
Dont like the GPL? Then dont link code to it. Dont like M$ word? Then don't use it. Its not like you dont have a choice. If you're forced to use linux, that isnt using the 'viral' properties of GPL, thats just using the software, which isnt forbidden under any GPL term (except the one about if the license clashes with law etc.. and perhaps not even then). If somebody sends you a word document, try to open it in some other program that can read (some) word documents, and if not, tell the other user to get a clue (politely of course), and send you the document in rtf or plain old text, or even just print it to a file (if they have a postscript printer, you have it in postscript.. but I'm getting away from the subject here).
The main point here is that you arent forced into doing something you dont want. Alternatives are out there (LaTeX, BSD license (and code under that license), other word processors, other licenses - I just picked those two because theyre the ones I like, and I know LaTeX isnt strictly a word processor, but it does the job I need a word processor to do)
(As an aside, I would love to have a nice alternative to say.. the ebola virus:), say maybe, the luck virus? (good variety of course))
One of the things that XML (to me) seems really suited for is what a lot of hype made it out to be in its early days: a document format for web pages.
Nowadays everyone claims the XML should be machine processed, and rearely required to be written by humans, but I would love to just be able to type something like:
and then have a CSS file apply styles to each of the tags as required, and just display it.
As it is now, with a bit of XSL to convert <tagname> to <div class="tagname> and wrap the document with surrounding html tags, and using.name { style: whatever; } instead of name {style: wahtever;} (i.e. stick a dot before to change from tagname to class entry) in the stylesheets, pretty much the same effect can be achieved.
The advantage if this is all documents are validated as a requirement of being displayed.. no more invalid html because its not possible. (Yes I know about xhtml, and its very nice - at least xhtml 2 is, but if browsers were forced to choke on invalid data, then in the process of testing web page display - people do test their pages dont they? - they would discover an error and correct it).
And of course its then easy to convert the data into any other xml based format with a couple more stylesheets.
My only problem is that I've not found a decent stylesheet parser that will just take a file, run it through one or more stylesheets, and display it, that will run over cgi, and not require any weird libraries to be installed, or XXX version of php that my isp doesnt happen to run, and allow me to say something like 'yeah, but before you transform it, just include this xml file here for a header and footer' But then again, I havent looked amazingly hard.
when I replace my current phone (in a few months - how nice of the phone company to give me a new phone for free). I'm thinking of leaving it connected to my computer as a dial-in modem (use a pay as you go sim, so there are no running costs), and have remote control over gsm for the pc.
Of course, this cant be done with all phones, just those with a GSM modem, but in a few months that could very well be the norm.
Another nice idea for a phone connected to the pc would be to get the pc to respond to incoming sms messages, which is doable on some phones I know of (ericsson t68(i) can have messages read via bluetooth/ir, but of course, just using the t68i as a bluetooth remote would work).
Hmm.. if theres any available, I guess you could even run a wap gateway on your computer when dialling in, for say computer status info etc.. DIY mobile isp.
I meant it would be some weird programming that depended on using add (thereby preserving the carry bit from the previous instruction for the next one), as opposed to adc. Of course, as soon as I mentioned weird I had a feeling that it had probably already been done, and is probably even common.
From what I've seen, the 5600 is very similar to the 5500, which has both sd/mmc and cf slots. The first thing I bought with my 5500 was a 256MB cf card and 128MB sd, storing programs on the sd card and mp3s (well oggs) on the compact flash, which I could remove and replace with the ethernet card when needed net access.
The real crippling factor is that in the 5500, the memory was all SDRAM which meant that you could flash with a different rom (openzaurus maybe) and have the full 64MB of main memory for loaded program use, and the SD card for storage. With the 5600 its 32MB SDRAM and 32MB flash, which sort of kills this idea. And if you ever want to run more than one app on the pda with 32MB, you quickly run out of memory.
The improvement in battery life is the only reason I would want to upgrade to a 5600, but even then you can get battery packs for the 5500, and its a LOT cheaper. (I paid £300 in january and its gone down to approx £200, probably even less now:( )
You missed:
- Coffee beans/grinder/caffetiere
- Sledgehammer
- Padlock to lock family members in another room when they start screaming at you for opening the pc. "YOU'RE GONNA BREAK IT!! WHAT ARE YOU DOING!!!"
Of course, you could use the sledgehammer instead of the padlock, or improvise with some rope, but all 3 of those are VERY important tools.
Tech guy 1> Hey, look at my cool new web based speech thingy that lets 1000's of users web pages talk to them! Tech guy 2> Bah.. bet it wouldnt support 2 people Tech guy 1> It would! Tech guy 2> Prove it... (loud musical sound of doom follows) post it to slashdot Tech guy 1> Ulp... (reluctantly taps away on the keyboard)
5 minutes later, strained sounds can faintly be heard from the smoking pile of rubble that used to be the server room, and the fried piece of circuit board that used to be the shiny new voice system crackles begin to wane, still trying to come up with 500,000 convincing renditions of "goatsec"
Oh we have restrictions.. or used to, now they just restrict your access if you are found to be running kazaa etc.. But for say just getting large iso's, they dont bother.
Actually inernet access is a specific extra charge of £92 (about $140) for the year, which is optional, and a lot cheaper than any DSL offering around (cheapest that doesnt impose stupid download limits is about £25-30 a month), and mobile internet access is unbelievably extortionate.. last I checked it was £5/meg for gprs (and I think it still is).
But my main point is, while £92 might still seem a considerable amount, this is for approx 100-400kbyte/sec access (and when they upgrade the line into university next year, somewhere in the region of 6MBytes/sec on a good day). Comparing this to 512kbit down/256kbit up for twice the price the university looks pretty cheap.
You'd be more likely to get the latter. Something tells me that outside of university dorms, internet access is going to be overpriced for a long long time.
I stand corrected. Life would be much easier though if 1 2 4 8 16... 1048576....were the only numbers. I think its time we made a nice shiny new standard to this effect.
Heh.. I've forgotten all about 8-bit architectures (and never knew anything about x86 anyway, which I'd assume is the same idea) - after most recently been readin about ARM assembly it just never occurred to me that it would be more costly than adding one more opcode to the instruction set.
What? Bitshifts _are_ the only multipication/division instructions worth considering!
Seriously, I suspect that combined with the add (sorry adc) instruction.. (and sub, add a two's complemented negative number), pretty much any multipication/division can be performed, and probably not much slower (and in a lot of cases faster) than a proper multiply instruction.
Floating point is a completely different matter though.
Slashdot Mirror
I missed this the first time round and probably should have posed this as a question, but its a bit late now.
Basically, what does Reason offer over running gcc -Wall -pedantic -anything else, running lint on the code, and then running the code with something like valgrind to detect dynamic memory problems? Okay, its a single tool as opposed to many of these and probably produces nicer reports, but what I think I'm really asking is, for the poor student like me, is there anything similar/have I missed out any really useful tools from the above (lint/valgrind/gcc with all warning options)?
OpenXFree86, and concentrate on proactive security whilst still being easy to use?
see 'shell company' and start formulating a reply along the lines of.. 'yeah, theres plenty, but few will let you run eggdrops/bnc servers.. etc..etc..'?
Wow.. that sounds amazing, especially considering its computer generated. So now I have to tell everyone that a computer can not only beat me at board games, but it can improvise better than I can as well :(
:)
Admittedly, jazz is a lot more random than other styles, and the chords for some (all?) of the songs were predetermined, but still, its the first computer generated music I've heard that didnt sound 'weird', unlike the ones from the toplevel article here, parts of which sound like they would make really good intros/backings to a tune, but needs something to make the music more melodic (I guess that where the GA comes in).
I'd be interested to know how many generations it takes for most of the tunes, and what fitness function he used (or if it was just.. listen to several and pick the best few).. guess its reading time
Or people who have their preferences set to show everything on the front page, which is a bitch when I forget to log in.
Isn't OpenBSD in canada because thats where Theo De Raadt lives?
spam comes from tins.
I have no idea why they have that daft pig icon for spam stories though.
I'd argue that whether it is viral isnt (or needn't be) the issue here.
:), say maybe, the luck virus? (good variety of course))
Both word documents and the GPL license prevent you from making use of {documents|code} in certain circumstances. (Yes I know the GPL grants you extra rights on copyright, but all licenses have conditions, thereby preventing use).
But the important thing is - YOU DO NOT HAVE TO USE EITHER in your work.
Dont like the GPL? Then dont link code to it. Dont like M$ word? Then don't use it. Its not like you dont have a choice. If you're forced to use linux, that isnt using the 'viral' properties of GPL, thats just using the software, which isnt forbidden under any GPL term (except the one about if the license clashes with law etc.. and perhaps not even then). If somebody sends you a word document, try to open it in some other program that can read (some) word documents, and if not, tell the other user to get a clue (politely of course), and send you the document in rtf or plain old text, or even just print it to a file (if they have a postscript printer, you have it in postscript.. but I'm getting away from the subject here).
The main point here is that you arent forced into doing something you dont want. Alternatives are out there (LaTeX, BSD license (and code under that license), other word processors, other licenses - I just picked those two because theyre the ones I like, and I know LaTeX isnt strictly a word processor, but it does the job I need a word processor to do)
(As an aside, I would love to have a nice alternative to say.. the ebola virus
OpenBSD isnt vulnerable :P
One of the things that XML (to me) seems really suited for is what a lot of hype made it out to be in its early days: a document format for web pages.
/>
.name { style: whatever; } instead of name {style: wahtever;} (i.e. stick a dot before to change from tagname to class entry) in the stylesheets, pretty much the same effect can be achieved.
Nowadays everyone claims the XML should be machine processed, and rearely required to be written by humans, but I would love to just be able to type something like:
<blogpost>
<name>mivok</name>
<date year="2003" month="3" day="18"
<subject>XML sucks</subject>
<body>
Blah Blah
</body>
</blogpost>
and then have a CSS file apply styles to each of the tags as required, and just display it.
As it is now, with a bit of XSL to convert <tagname> to <div class="tagname> and wrap the document with surrounding html tags, and using
The advantage if this is all documents are validated as a requirement of being displayed.. no more invalid html because its not possible. (Yes I know about xhtml, and its very nice - at least xhtml 2 is, but if browsers were forced to choke on invalid data, then in the process of testing web page display - people do test their pages dont they? - they would discover an error and correct it).
And of course its then easy to convert the data into any other xml based format with a couple more stylesheets.
My only problem is that I've not found a decent stylesheet parser that will just take a file, run it through one or more stylesheets, and display it, that will run over cgi, and not require any weird libraries to be installed, or XXX version of php that my isp doesnt happen to run, and allow me to say something like 'yeah, but before you transform it, just include this xml file here for a header and footer'
But then again, I havent looked amazingly hard.
when I replace my current phone (in a few months - how nice of the phone company to give me a new phone for free). I'm thinking of leaving it connected to my computer as a dial-in modem (use a pay as you go sim, so there are no running costs), and have remote control over gsm for the pc.
Of course, this cant be done with all phones, just those with a GSM modem, but in a few months that could very well be the norm.
Another nice idea for a phone connected to the pc would be to get the pc to respond to incoming sms messages, which is doable on some phones I know of (ericsson t68(i) can have messages read via bluetooth/ir, but of course, just using the t68i as a bluetooth remote would work).
Hmm.. if theres any available, I guess you could even run a wap gateway on your computer when dialling in, for say computer status info etc.. DIY mobile isp.
I meant it would be some weird programming that depended on using add (thereby preserving the carry bit from the previous instruction for the next one), as opposed to adc.
Of course, as soon as I mentioned weird I had a feeling that it had probably already been done, and is probably even common.
Yeah, spotted it after I posted, but it still doesnt make up for (imho) halving the amount of sdram (even if it wasnt usable in the default sharp rom)
Theres also The zaurus DevNet forums geared mostly towards developers, but still very useful.
From what I've seen, the 5600 is very similar to the 5500, which has both sd/mmc and cf slots. The first thing I bought with my 5500 was a 256MB cf card and 128MB sd, storing programs on the sd card and mp3s (well oggs) on the compact flash, which I could remove and replace with the ethernet card when needed net access.
:( )
The real crippling factor is that in the 5500, the memory was all SDRAM which meant that you could flash with a different rom (openzaurus maybe) and have the full 64MB of main memory for loaded program use, and the SD card for storage. With the 5600 its 32MB SDRAM and 32MB flash, which sort of kills this idea. And if you ever want to run more than one app on the pda with 32MB, you quickly run out of memory.
The improvement in battery life is the only reason I would want to upgrade to a 5600, but even then you can get battery packs for the 5500, and its a LOT cheaper. (I paid £300 in january and its gone down to approx £200, probably even less now
You missed:
- Coffee beans/grinder/caffetiere
- Sledgehammer
- Padlock to lock family members in another room when they start screaming at you for opening the pc.
"YOU'RE GONNA BREAK IT!! WHAT ARE YOU DOING!!!"
Of course, you could use the sledgehammer instead of the padlock, or improvise with some rope, but all 3 of those are VERY important tools.
Okay.. I dont have mod points to mod you up, so I'll have to reply instead.
Half the family/friends tech support calls I get are...
'I can't get it on the internet! Make it work'
Tech guy 1> Hey, look at my cool new web based speech thingy that lets 1000's of users web pages talk to them!
Tech guy 2> Bah.. bet it wouldnt support 2 people
Tech guy 1> It would!
Tech guy 2> Prove it... (loud musical sound of doom follows) post it to slashdot
Tech guy 1> Ulp... (reluctantly taps away on the keyboard)
5 minutes later, strained sounds can faintly be heard from the smoking pile of rubble that used to be the server room, and the fried piece of circuit board that used to be the shiny new voice system crackles begin to wane, still trying to come up with 500,000 convincing renditions of "goatsec"
Oh we have restrictions.. or used to, now they just restrict your access if you are found to be running kazaa etc..
But for say just getting large iso's, they dont bother.
Actually inernet access is a specific extra charge of £92 (about $140) for the year, which is optional, and a lot cheaper than any DSL offering around (cheapest that doesnt impose stupid download limits is about £25-30 a month), and mobile internet access is unbelievably extortionate.. last I checked it was £5/meg for gprs (and I think it still is).
But my main point is, while £92 might still seem a considerable amount, this is for approx 100-400kbyte/sec access (and when they upgrade the line into university next year, somewhere in the region of 6MBytes/sec on a good day). Comparing this to 512kbit down/256kbit up for twice the price the university looks pretty cheap.
You'd be more likely to get the latter.
Something tells me that outside of university dorms, internet access is going to be overpriced for a long long time.
I stand corrected.
Life would be much easier though if 1 2 4 8 16... 1048576....were the only numbers. I think its time we made a nice shiny new standard to this effect.
Puts a whole new slant on frying circuitry.
Heh.. I've forgotten all about 8-bit architectures (and never knew anything about x86 anyway, which I'd assume is the same idea) - after most recently been readin about ARM assembly it just never occurred to me that it would be more costly than adding one more opcode to the instruction set.
What? Bitshifts _are_ the only multipication/division instructions worth considering!
Seriously, I suspect that combined with the add (sorry adc) instruction.. (and sub, add a two's complemented negative number), pretty much any multipication/division can be performed, and probably not much slower (and in a lot of cases faster) than a proper multiply instruction.
Floating point is a completely different matter though.