I'm curious whether the productivity increase you commented on was just an off the cuff "you might be better at Python" comment or an actual, documented speed increase that most programmers would be able to take advantage of.
Similar to the poster you were responding too, most of my programming these days is in the work environment, leaving little time and energy outside of that to work with new languages. While I do have enough python experience to know I love it, I have yet to write anything more complicated than an isometric map generator and partial editor (on the gui side) and the beginning of a data historian (on the pure code side). Day-to-day I work with C# and ASP.net and way too little time. If IronPython is more efficient to write code in (in a time:result ratio) than I definatly am interested.
There are a lot of posts that mention this and a lot that whine about not getting what they thought was advertised. For the record I do not and have not worked for an ISP.
ISP's do not advertise a fixed rate connection (at least not for your standard residential account like we are talking about here). They advertise a service that can burst up to a certain speed. I don't know if they still have floor guarantees anymore (ie, you will get at least 256Kb/s, bursting to 3Mb/s) but I know some companies did for a while.
What amazes me is that people expect to get something like the speed of multiple T-1's for $40/month.
This is how it works. You ar ean ISP and you buy some lines, say a T1. That T1 will support 1.54 Mb/s and you want to offer your users 256Kb/s connections. Now if you assume that those users will always be connectd and always maxing their pipe, that means you can only have 6 users. Doesn't mean much profit at the end of the day (like negative). However there is a flaw here, your standard home user is not going to max their pipe and won't be on at high loads 24/7. Some will be at evening jobs, some will be watching TV instead, some will just be browsing web pages at a click per few minutes. So what we do is come up with a ratio of how many people will be idle versus actually online. Lets say it's 8:1 (some companies go much higher). That would mean you could then increase your proposed customer pool to 48 (8*6). Could all 48 people login all at once? Nope. Will they all be likely to get an acceptable level of service? Yes, especially since we assumed that 1:8 users would be logged in at any given point in time, but didn't work out what percentage of their bandwidth they would consume on average.
This is a very simplified version of the process. In most cases companies do not use customer's maximum rate to calculate how large their pool can be, they use a value they believe to be an acceptable level of service or an average level of use for their users.
This is where your service problems come in. Once a company hit's their target, even if it is a fairly good estimate, they then have two choices, expand the number a little or expand their expenses and invest in more lines and hope the customers will come.
Oh, and another easy point: How many of you work at companies that rent the equivalent of a T1 for each of their employees? How many of you share it with 50 other users? 100 other users?
So your saying commenting/version-tracking does work for small groups of people? Thats been my experience as well. Sharepoint has a lot of promise in extending these, but unfortunatly I think Sharepoint has fallen fairly wide of the mark in terms of usefulness and useability. I believe that the granparent's point was that this feature (commenting/version-tracking) is useful in some circumstances and is not included in Open Office. And I am willing to agree that it is useful in many cirucumstances. However: OOo Writer -> Tools -> Options -> Text Document -> Changes And of course File -> Versions for saving new versions of a file.
I haven't actually used that option in OOo before (maybe because I never noticed it among the other 100 options you can change in that menu), so I couldn't say if it works beter or worse then the color-coded bubbles and underline system that Word uses. But it is in there.
Yep, we're just waiting for those self-inflating cars that can grow exponentially to stay ahead of the size of other cars on the road. That way everyone with a below-median car can consistently "grow" their car to move the the top group, at which point there is a new group below the "median" that will inflate their cars higher, and so on.
Unless you were talking about median cars world-wide. Although if that were the case you have a long time to wait before our larger-than-median cars start colliding with drivers in Germany, China, Australia, etc.
And if you were talking about Americans as in "citizens of the USA" than we can all show a little relief. Just think how bad those intercontinent collisions would be if, say, the citizens of Canada, Mexico, and South American countries started getting into the act...
Either that or the person responsible for that comment is one of the cowboy coders, for whom any non-coding time is seen as a waste (ie, testing, retesting, documentation, etc).
Ah man, now I really want to put some watercooling on my old P4 rig. I mean, the 30% overclock is nothing compared to a 0.2% increase...
P4 1.6a that has been running at a 30% overclock on aircooling for 3 years. It gets maybe a few hours off per year and has so far outlived a motherboard and PSU and is still running strong. Thanks for playing...
At the same time, aren't all of the reasons you mentioned also good reasons to seperate the SQL from the code asmuch as possible? Move everything to stored procedures, perhaps even encapsulate those calls in your code in their own functions, etc and then you can make just about any major changes ot the backend you want, provided you also update the stored procs. From the frontend you can change to any other type of datasource you want, provided you update the functions that are actually getting your data.
There is some logic that is easily moved to the database layer that does not need to be part of the applicaiton logic. I have seen web apps that do running totals or category totals completely on the fly when a simple SUM statement would have sufficed. Logic for INSET vs UPDATE statements canactually be handled by the database in some circumstances, as it actually allows greater flexibility in the future. Want to add traceability? If you already took the previous step than this now bcomes trivial because it can be done in the database withoutever touching the application code.
Same thing with the front-end. Since we are talking web, there are many reasons (beyinf eficiency) to keep your HTML sparse and try to place most of the formatting and prettiness in external CSS files. This does add complexity if you judge complexity by number of languages, but it adds simplicity to maintainability. Need to make vast sweeping changes to the color of somehting on the site? No problem, you don't need to know PHP/ASP/JSP/whatever, just modify the CSS file (if it was done right). You still have some crossover when it comes to the server-side code that is operating behind the scenes, as you do need o know CSS, HTML, and whatever server-side language your using, but what you won't be doing is searching through who knows how much code for the single remaining bgcolor attribute or inline style attribute.
I agree that taking this concept too far is worse than no seperation at all. But in moderation you do get a large number of advantages. I actualy had a good example of "too far as to be useless" the other day when our coop explained the restricionts their teachers were putting on them for a software project: 5 distinct layers in a web app, not counting CSS, with every layer bleading into at least 2 more. A change to the database directly affected a minimum of 2 other layers as the hardcoded SQL statements were in the next layer and the table-specific classes were in an additional layer (relationships were variables references between the objects)...there was more, but it bothers me everytime he mentions it.
I will be one of the few posters (that I have sen so far) to say that I have poor grammar and several words I consistently mangle. For whatever reason my brain continues to spit out the wrong spelling for several common words and my grammar skills are likely to be barely adequate for high school. However, there is a great difference between the occasional spelling or grammar error and the travesties often posted requesting aid on the forums. What I have often observed is that these same people that can barely fashion a sentance will often have the same difficulty fashioning clean, easy to read code. There will be the occasional writer that tries to squeeze 4 or 5 sentances worth of information into a single sentace without even as much as a period, but can manage to write clean looking code, but these are generally the exception. What is unfortunate is the one or two people, as the parent poster mentioned, that knew so little English or received incredibly broken translations from the translation tool they were using, that we were incapable of assisting them. In a way I almost felt as if they deserved our assistance more than many of the native English speakers that could not be bothered to hit the comma, period, or even shift key while typing their request.
Then again, the worse a question is typed, the more inclined I am of giving a one word answer. Perhaps this issue is self-correcting at some point.
A friend of mine has his machine setup with SageTV and Jinzora music server. The music server allows him to listen to music right there (in jukebox mode) or stream it to any other computer in the house. His expensive 200-disc changer became a birthday present for a family member. The SageTV install means that he can now record soaps for his fiance and let her watch them when he isn't home.:) (as well as the usual timeshifting capabilities, recording things for himself, etc). He could open the server to the whole world and let them stream music, but tat wouldn't exactly be legal or leave him any bandwidth, butheis putting in a new gateway router and is going to setup a VPN tunnel from work so we can stream music to work (automatically transcodes from the original flac files to mp3/ogg/whatever). His fiance recently got an iPod, guess how happy she is to have their entire CD collection available in one location without having to hunt through a disc changer/cd book/basement for the orginal CD? Since I am using MythTV on mine, I get the additional advantage of having a couple games on it to keep my fiance and other visitors busy, such as Mario 3, mario Kart, etc for SNES.
Excellent, if I had mod points I would have burned some. I completely agree. Everyday some new set of articles is posted about "Web2.0", etc and yet things like AJAX don't seem to be revolutionary as much as small steps and, in most cases, gap fillers. I've written some "AJAX" stuff in the past. Granted only a few of sites I wrote using javascript and asynchronous XML calls were written after there was a name for it, but as the parent poster said, it is not revolutionary. There are some sites out there that are heavily AJAX-ified. Some of them are useful, some of them are just feature-filled for features sake. In most of the cases I have seen, though, AJAX has been used to fill the gaps, as a polish, if you will. Now don't get me wrong, this is not a bad thing, but even GMail and Google maps are far from the promised revolution we read about in articles such as this.
All this AJAX "revolution" does for us is allow us to treat the frontend web page as a bit more of a client, as opposed to treating it as content in a thin client. Flash did as much, if not more. The only difference here is that AJAX isn't a platform that requirs a plugin, it is instead (at it's core) a group of existing plugins accessible by the browser. How long has AJAX been around as the cool, hip thing? It's already available to every developer with notepad/vi and a browser, it's not very complicated to implement, so where is this revolution that has been remarked on since the day it got an acronym all it's own?
And a commentary on the original article: I lost interest in reading it right after the author said that our OS's are out of date and what we really need is a web-based desktop to truly leverage all of the fantastic capabilities of our machines...right. I didn't look too closely, but to me it just looked like he was selling start pages, similar to what Yahoo and many others have had for ages...personally I like my desktop environment to do a little more.
Oddly enough I get the same popup message when I start the game on my 7800GTX. I didn't get the message on my ATI 9700 Pro. Granted my 7800GTX drivers are out of date because I don't like running bleeding edge when I'm already getting a 90+ framerate (my ATI were up to date because 25fps was painful), but I'm willing to bet that a large part of the problem is just the fact that BF2 is a beast. My older windows box would consistently gain 10fps for several days after a defrag (avg of 35fps), only to fall off with time to an avg in the low 20's. A total uninstall, defrag, reinstall, defrag actually improved it by about 15-20fps from the lower average for a few days. HL2 looked beautiful.
So basically what I'm saying is that BF2 may not be the best candidate for comparison. It's got a boatload of outstanding bugs, is a beast when it comes to hardware requirements, and seems to have some odd scaling issues with certain chipsets.
Part of the problems faced in the first job out is the deadline. Yes, you had deadlines by the dozen in school, and each major one could potentially ruin several months of your life if missed, however real world work doesn't have to be like that. At least not to the same degree.
First of all, your not being assigned projects or tasks that have been carefully balanced to fit in an allotted time with a nice set of requirements. Your probably getting requirements that float all over the map and a "we need it yesterday" deadline. I sympathise, I have several projects in the air right now that are similar. However, the bonus on these projects is that you can apologize, tell the customer how much you want to have it done on time, and push out the date because such-and-such project takes priority, or because you "really want to give it the time to do it right". Better yet, if you have a good manager, they should be doing it. You have to manage your load instead of allowing it to manage you. Don't be afraid to sit down with your manager and talk about relative priorities of projects and which deadlines seem unhealthy. Remember, your customer wants a working solution in an impossible amount of time. You can either throw something together in a short time and take one step closer to mental break down or get your manager to help you make a realistic deadline that he can then go back and explain to the customer. It's in everyone's best interests that you don't have a mental break down and that you don't slap everything together as fast and dirty as you can.
Second, I had no second:P
In any case, good luck with your situation, been there, or somewhere closely approximating there.
I had been planning on trying Myth for a while and just got around to doing so. It went extremely smoothly for the most part. I did have some minor isues with the haupauge remote (buttons in LIRC config file needs to be remapped, not a big deal) and my GeForce 2 that I was trying had some issues that would have been solved a lot quicker if I had noticed the NTSC jumper on the card:P
I'm using the KnoppMyth distro. As long as you have an internet connection when you install, and enter the right time difference in the setup screen before downloading the listings, it's fairly quick. Your haupauges should work out of the box. You didn't mention your video card, so I wouldn't know. As long as you pick an Nvidia you should be good. I'm planning on picking up an Nvidia 6200 (fanless and capable of HD) for $53 for video out. Right now my CPU maxes at about 22% usage with simultaneous recording on the 350 and playback to the antiquated GF2 MX400.
Plus I'm looking forward to the next release, as it looks like they will be making the MAME emulator section a lot more dynamic as well as including an N64 emulator. My MS Sidewinder pad was awesome last night when I was playing the SNES Mario Kart last night:P
Unless you planned on sticking an RFID tag to your forehead and masquerading as a crate/box/skid in the closest loading dock/manufacturing facility/etc then I think you will be ok.
Re:Single component need for full realization
on
The Year of the HTPC
·
· Score: 1
ATI OCUR? Slated for 2006?
The fact that ATIis doing it should drive a lot more companies into rushing to get their own put together as well.
The KDE app DigiKam allows nested tags. I haven't found out if you can do searches based on "has this tag but resides somewherr under this other tag" but it does allow me to orgranize my tags hierarchally as well as having a physical folder structure at the same time. The tags act more like a cross-referencing system rather than an organizational structure. I kind of like it better. I actually like it better. The folders are specific download dates, generally allowing me to tag the entire contents as a certain event (minus the 3 or 4 random ones) then I go back through and tag it with all the people/pets/places/etc that are in the photo. I guess the folder structur isn't necessary, but it helps agreat deal with moving old photo directories that have pictures with no EXIF info encoded in them.
Concerning the word customer: You missed the second sentance there. I don't really care for the usage, but calling users, etc "customers" has been common in a lot of circles for a little while now. I was just pointing out that under that more general definition, nearly all software is customer-driven (if we don't include ourselves for personal projects). If it makes you feel more comfortable, then replace the phrase customer-driven with user-driven. I would hope that you would agree that most software is indeed user-driven, even if the intended user is oneself.
I still disagree that writing a portion of a system allows the writer to claim that they were the creator of the larger, pre-existing system that there piece is being used in. Even if we are talking about a piece of software that is actually public domain with a completely open copyright, I could not consider myself as having created a piece of software after I download it and add a few new components. I may have created a collection of pre-existing software or pre-existing software and my own contributions, but I cannot claim that I am the creator of the original software.
I do not disgaree with your last statements. I use FOSS software and libraries and I also use proprietary software. Where I disagree the most is with Stallmans comments that the original poster was disagreeing with:
JP: But is there still an ethical issue if there is no alternative? If, say, there is no free software way of doing a particular job, for ZNet for example? RMS: One can live without doing those jobs.
This and comments similar to this are part of the issue I have. I have software problems that aren't even close to being solved by available FOSS software. heck, I have problems that the available proprietary software is about 4-5 years behind on. Should I sink $200k-$500k into custom development (in either cash or time) of manufacturing software that I need or should I depend on cheaper, proprietary solutions that can scale to new devices and new sizes for a mch lower price tag? I choose to go with proprietary software, but according to Stallman I should have spent the money/time or consider it a job not worth doing. Sorry for the long and somewhat general example, but wanted to clear up what I was supporting from the original parent poster.
yeah, sorry, that wasn't actually a commentary on your post's meaning, I was just pointing out that the "shortage of unwanted kids" struck me as a funny phrase:)
I'm not necesarally disagreeing or agreeing with you, just disagreeing with your points.
By "real" job, I mean a job in which you are required to meet deadlines imposed by customers and to produce end products specified by customers So you're clearly excluding almost every one...
Do you work in the real world? IS/IT Developers develop (or find and configure) software for their customers (ie, other people who work for the company). Developers at software houses are even more obvious. Network admins, systems admins, dba's, etc all generally have deadlines to produce the goods to their 'customers', whether they be actual customers, coworkers, or other departments in their company/organization. Tech support does it. Hell, I did it in a car for five years delivering pizza. The word "Customer" covers a lot more than it's literal definition these days. I may not care for that definition much, but I do have to admit that it exists and is in popular use.
AFAIK, most software projects are not customer-driven. Please provide links.
Care to provide numbers? In House development is generally customer driven, for sale software development is customer driven, etc. Many open source projects may be just a guy/gal with no deadlines and no expectations coding a project, but the big ones have release deadlines, user groups, bugs to fix, etc. those users are just as much customers as the people in the marketing department that have decided they need a brand new website.
Their track record is simply appalling. Their track record is simply amazing. In a few years, they have come with a major complete operating system that is a direct competitor of Microsoft and Apple.
I once built a driver, therefore I will take take credit for having written Windows. The point being that writing a portion of something used in a system does not make a developer or organization the maker of said system. The word for that is contributor. Or developer of a seperate piece of software.
only an idiot would pay GNU to write it for them, bcos they've already proved their inability to do work to time. You mean Microsoft?
It's unfortunate, but there is no argument here to pick apart. Rather than supply an actual argument you fall back on poking the finger at Microsoft. Good job. And note that before you point your finger at my last point, the comment I made was in direct support of an argument, not a standalone random statement.
---
I think the original author did have some valid points and in a couple cases took some arguments a little far. I agree with several other posters that have pointed out that Linux would not be what it is today without GNU tools/software. But I also agree with the original poster that most of what I have read on Stallman gives me the feeling of an academic. I still recall randomly running into a mailing list for a project he was working on. For the hell of it I started reading. He spent days trying to hold back a pre-release beta because there were still outstanding bugs. The purpose of that particular pre-release beta was to gather more bug information and get real users using it to gather that information. Instead he wanted to wait until there were no (some class) bugs in it before doing the pre-release. Most of his responses came off as childish.
I support the idea of open source and if I had 4-6 more hours in the day I would probably contribute more, as it is I have only made modifications to existing projects rather than take on the responsibility of being a team member or such for one. But I have to say I don't actually like RMS. I find him to be somewhere between a whiny kid and a wild-eyed, "The End Is Nigh" madman. Someone that inspires the fans (and fanatics) that already exist, but doesn't provide the non-fans with a lot of inspiration beyond, maybe, sidling away a little to give him space, just in case
Slashdot Mirror
I'm curious whether the productivity increase you commented on was just an off the cuff "you might be better at Python" comment or an actual, documented speed increase that most programmers would be able to take advantage of.
Similar to the poster you were responding too, most of my programming these days is in the work environment, leaving little time and energy outside of that to work with new languages. While I do have enough python experience to know I love it, I have yet to write anything more complicated than an isometric map generator and partial editor (on the gui side) and the beginning of a data historian (on the pure code side). Day-to-day I work with C# and ASP.net and way too little time. If IronPython is more efficient to write code in (in a time:result ratio) than I definatly am interested.
Plus I love my list slicing...
There are a lot of posts that mention this and a lot that whine about not getting what they thought was advertised. For the record I do not and have not worked for an ISP.
ISP's do not advertise a fixed rate connection (at least not for your standard residential account like we are talking about here). They advertise a service that can burst up to a certain speed. I don't know if they still have floor guarantees anymore (ie, you will get at least 256Kb/s, bursting to 3Mb/s) but I know some companies did for a while.
What amazes me is that people expect to get something like the speed of multiple T-1's for $40/month.
This is how it works. You ar ean ISP and you buy some lines, say a T1. That T1 will support 1.54 Mb/s and you want to offer your users 256Kb/s connections. Now if you assume that those users will always be connectd and always maxing their pipe, that means you can only have 6 users. Doesn't mean much profit at the end of the day (like negative). However there is a flaw here, your standard home user is not going to max their pipe and won't be on at high loads 24/7. Some will be at evening jobs, some will be watching TV instead, some will just be browsing web pages at a click per few minutes. So what we do is come up with a ratio of how many people will be idle versus actually online. Lets say it's 8:1 (some companies go much higher). That would mean you could then increase your proposed customer pool to 48 (8*6). Could all 48 people login all at once? Nope. Will they all be likely to get an acceptable level of service? Yes, especially since we assumed that 1:8 users would be logged in at any given point in time, but didn't work out what percentage of their bandwidth they would consume on average.
This is a very simplified version of the process. In most cases companies do not use customer's maximum rate to calculate how large their pool can be, they use a value they believe to be an acceptable level of service or an average level of use for their users.
This is where your service problems come in.
Once a company hit's their target, even if it is a fairly good estimate, they then have two choices, expand the number a little or expand their expenses and invest in more lines and hope the customers will come.
Oh, and another easy point: How many of you work at companies that rent the equivalent of a T1 for each of their employees? How many of you share it with 50 other users? 100 other users?
Sounds like you want an OQO...or maybe an OQO on clearance sale :P
So your saying commenting/version-tracking does work for small groups of people? Thats been my experience as well. Sharepoint has a lot of promise in extending these, but unfortunatly I think Sharepoint has fallen fairly wide of the mark in terms of usefulness and useability.
I believe that the granparent's point was that this feature (commenting/version-tracking) is useful in some circumstances and is not included in Open Office. And I am willing to agree that it is useful in many cirucumstances. However:
OOo Writer -> Tools -> Options -> Text Document -> Changes
And of course File -> Versions for saving new versions of a file.
I haven't actually used that option in OOo before (maybe because I never noticed it among the other 100 options you can change in that menu), so I couldn't say if it works beter or worse then the color-coded bubbles and underline system that Word uses. But it is in there.
Yep, we're just waiting for those self-inflating cars that can grow exponentially to stay ahead of the size of other cars on the road. That way everyone with a below-median car can consistently "grow" their car to move the the top group, at which point there is a new group below the "median" that will inflate their cars higher, and so on.
Unless you were talking about median cars world-wide. Although if that were the case you have a long time to wait before our larger-than-median cars start colliding with drivers in Germany, China, Australia, etc.
And if you were talking about Americans as in "citizens of the USA" than we can all show a little relief. Just think how bad those intercontinent collisions would be if, say, the citizens of Canada, Mexico, and South American countries started getting into the act...
So basically what your saying is that you not only didn't RTFA, but you made a new record by posting without even reading the summary either?
Either that or the person responsible for that comment is one of the cowboy coders, for whom any non-coding time is seen as a waste (ie, testing, retesting, documentation, etc).
Ah man, now I really want to put some watercooling on my old P4 rig. I mean, the 30% overclock is nothing compared to a 0.2% increase...
P4 1.6a that has been running at a 30% overclock on aircooling for 3 years. It gets maybe a few hours off per year and has so far outlived a motherboard and PSU and is still running strong. Thanks for playing...
It must be the caes. I was thinking the same thing. I spent $1200 for new parts to rebuild this box and it is barely audible from a couple feet away:
Antex P180 Case
Enermax NoiseTaker PSU
Athlon64 3700+
Stock heatsink - ditton on the source of noise
BFG 7800 GTX OC
Matched 1GB Patriot RAM sticks
200GB Seagate SATA
SATA DVD+-R (can't remember vendor)
Epox mobo
As the parent said, it isn't silent, but I have to look at it to tell if it's on if a game is playing.
At the same time, aren't all of the reasons you mentioned also good reasons to seperate the SQL from the code asmuch as possible? Move everything to stored procedures, perhaps even encapsulate those calls in your code in their own functions, etc and then you can make just about any major changes ot the backend you want, provided you also update the stored procs. From the frontend you can change to any other type of datasource you want, provided you update the functions that are actually getting your data.
There is some logic that is easily moved to the database layer that does not need to be part of the applicaiton logic. I have seen web apps that do running totals or category totals completely on the fly when a simple SUM statement would have sufficed. Logic for INSET vs UPDATE statements canactually be handled by the database in some circumstances, as it actually allows greater flexibility in the future. Want to add traceability? If you already took the previous step than this now bcomes trivial because it can be done in the database withoutever touching the application code.
Same thing with the front-end. Since we are talking web, there are many reasons (beyinf eficiency) to keep your HTML sparse and try to place most of the formatting and prettiness in external CSS files. This does add complexity if you judge complexity by number of languages, but it adds simplicity to maintainability. Need to make vast sweeping changes to the color of somehting on the site? No problem, you don't need to know PHP/ASP/JSP/whatever, just modify the CSS file (if it was done right). You still have some crossover when it comes to the server-side code that is operating behind the scenes, as you do need o know CSS, HTML, and whatever server-side language your using, but what you won't be doing is searching through who knows how much code for the single remaining bgcolor attribute or inline style attribute.
I agree that taking this concept too far is worse than no seperation at all. But in moderation you do get a large number of advantages. I actualy had a good example of "too far as to be useless" the other day when our coop explained the restricionts their teachers were putting on them for a software project:
5 distinct layers in a web app, not counting CSS, with every layer bleading into at least 2 more. A change to the database directly affected a minimum of 2 other layers as the hardcoded SQL statements were in the next layer and the table-specific classes were in an additional layer (relationships were variables references between the objects)...there was more, but it bothers me everytime he mentions it.
They may have forgotten it, but it's still in use :(
Agreed, double and treble.
I will be one of the few posters (that I have sen so far) to say that I have poor grammar and several words I consistently mangle. For whatever reason my brain continues to spit out the wrong spelling for several common words and my grammar skills are likely to be barely adequate for high school. However, there is a great difference between the occasional spelling or grammar error and the travesties often posted requesting aid on the forums. What I have often observed is that these same people that can barely fashion a sentance will often have the same difficulty fashioning clean, easy to read code. There will be the occasional writer that tries to squeeze 4 or 5 sentances worth of information into a single sentace without even as much as a period, but can manage to write clean looking code, but these are generally the exception.
What is unfortunate is the one or two people, as the parent poster mentioned, that knew so little English or received incredibly broken translations from the translation tool they were using, that we were incapable of assisting them. In a way I almost felt as if they deserved our assistance more than many of the native English speakers that could not be bothered to hit the comma, period, or even shift key while typing their request.
Then again, the worse a question is typed, the more inclined I am of giving a one word answer. Perhaps this issue is self-correcting at some point.
A friend of mine has his machine setup with SageTV and Jinzora music server. The music server allows him to listen to music right there (in jukebox mode) or stream it to any other computer in the house. His expensive 200-disc changer became a birthday present for a family member. :) (as well as the usual timeshifting capabilities, recording things for himself, etc). He could open the server to the whole world and let them stream music, but tat wouldn't exactly be legal or leave him any bandwidth, butheis putting in a new gateway router and is going to setup a VPN tunnel from work so we can stream music to work (automatically transcodes from the original flac files to mp3/ogg/whatever).
The SageTV install means that he can now record soaps for his fiance and let her watch them when he isn't home.
His fiance recently got an iPod, guess how happy she is to have their entire CD collection available in one location without having to hunt through a disc changer/cd book/basement for the orginal CD?
Since I am using MythTV on mine, I get the additional advantage of having a couple games on it to keep my fiance and other visitors busy, such as Mario 3, mario Kart, etc for SNES.
Excellent, if I had mod points I would have burned some. I completely agree. Everyday some new set of articles is posted about "Web2.0", etc and yet things like AJAX don't seem to be revolutionary as much as small steps and, in most cases, gap fillers.
I've written some "AJAX" stuff in the past. Granted only a few of sites I wrote using javascript and asynchronous XML calls were written after there was a name for it, but as the parent poster said, it is not revolutionary. There are some sites out there that are heavily AJAX-ified. Some of them are useful, some of them are just feature-filled for features sake. In most of the cases I have seen, though, AJAX has been used to fill the gaps, as a polish, if you will. Now don't get me wrong, this is not a bad thing, but even GMail and Google maps are far from the promised revolution we read about in articles such as this.
All this AJAX "revolution" does for us is allow us to treat the frontend web page as a bit more of a client, as opposed to treating it as content in a thin client. Flash did as much, if not more. The only difference here is that AJAX isn't a platform that requirs a plugin, it is instead (at it's core) a group of existing plugins accessible by the browser. How long has AJAX been around as the cool, hip thing? It's already available to every developer with notepad/vi and a browser, it's not very complicated to implement, so where is this revolution that has been remarked on since the day it got an acronym all it's own?
And a commentary on the original article: I lost interest in reading it right after the author said that our OS's are out of date and what we really need is a web-based desktop to truly leverage all of the fantastic capabilities of our machines...right. I didn't look too closely, but to me it just looked like he was selling start pages, similar to what Yahoo and many others have had for ages...personally I like my desktop environment to do a little more.
Oddly enough I get the same popup message when I start the game on my 7800GTX. I didn't get the message on my ATI 9700 Pro.
Granted my 7800GTX drivers are out of date because I don't like running bleeding edge when I'm already getting a 90+ framerate (my ATI were up to date because 25fps was painful), but I'm willing to bet that a large part of the problem is just the fact that BF2 is a beast.
My older windows box would consistently gain 10fps for several days after a defrag (avg of 35fps), only to fall off with time to an avg in the low 20's. A total uninstall, defrag, reinstall, defrag actually improved it by about 15-20fps from the lower average for a few days.
HL2 looked beautiful.
So basically what I'm saying is that BF2 may not be the best candidate for comparison. It's got a boatload of outstanding bugs, is a beast when it comes to hardware requirements, and seems to have some odd scaling issues with certain chipsets.
I agree.
:P
Part of the problems faced in the first job out is the deadline. Yes, you had deadlines by the dozen in school, and each major one could potentially ruin several months of your life if missed, however real world work doesn't have to be like that. At least not to the same degree.
First of all, your not being assigned projects or tasks that have been carefully balanced to fit in an allotted time with a nice set of requirements. Your probably getting requirements that float all over the map and a "we need it yesterday" deadline. I sympathise, I have several projects in the air right now that are similar. However, the bonus on these projects is that you can apologize, tell the customer how much you want to have it done on time, and push out the date because such-and-such project takes priority, or because you "really want to give it the time to do it right". Better yet, if you have a good manager, they should be doing it. You have to manage your load instead of allowing it to manage you. Don't be afraid to sit down with your manager and talk about relative priorities of projects and which deadlines seem unhealthy. Remember, your customer wants a working solution in an impossible amount of time. You can either throw something together in a short time and take one step closer to mental break down or get your manager to help you make a realistic deadline that he can then go back and explain to the customer. It's in everyone's best interests that you don't have a mental break down and that you don't slap everything together as fast and dirty as you can.
Second, I had no second
In any case, good luck with your situation, been there, or somewhere closely approximating there.
I had been planning on trying Myth for a while and just got around to doing so. It went extremely smoothly for the most part. I did have some minor isues with the haupauge remote (buttons in LIRC config file needs to be remapped, not a big deal) and my GeForce 2 that I was trying had some issues that would have been solved a lot quicker if I had noticed the NTSC jumper on the card :P
:P
I'm using the KnoppMyth distro. As long as you have an internet connection when you install, and enter the right time difference in the setup screen before downloading the listings, it's fairly quick. Your haupauges should work out of the box. You didn't mention your video card, so I wouldn't know. As long as you pick an Nvidia you should be good. I'm planning on picking up an Nvidia 6200 (fanless and capable of HD) for $53 for video out. Right now my CPU maxes at about 22% usage with simultaneous recording on the 350 and playback to the antiquated GF2 MX400.
Plus I'm looking forward to the next release, as it looks like they will be making the MAME emulator section a lot more dynamic as well as including an N64 emulator. My MS Sidewinder pad was awesome last night when I was playing the SNES Mario Kart last night
Unless you planned on sticking an RFID tag to your forehead and masquerading as a crate/box/skid in the closest loading dock/manufacturing facility/etc then I think you will be ok.
ATI OCUR? Slated for 2006?
The fact that ATIis doing it should drive a lot more companies into rushing to get their own put together as well.
Or better yet, RTF format so they can keep the chapter headings and font formatting the author used to denote thoughts and whatnot.
MS was already doing their magic on SOAP many years ago when all the first sets of libraries were fresh and new. They embraced XML a while ago.
The KDE app DigiKam allows nested tags. I haven't found out if you can do searches based on "has this tag but resides somewherr under this other tag" but it does allow me to orgranize my tags hierarchally as well as having a physical folder structure at the same time. The tags act more like a cross-referencing system rather than an organizational structure. I kind of like it better. I actually like it better. The folders are specific download dates, generally allowing me to tag the entire contents as a certain event (minus the 3 or 4 random ones) then I go back through and tag it with all the people/pets/places/etc that are in the photo. I guess the folder structur isn't necessary, but it helps agreat deal with moving old photo directories that have pictures with no EXIF info encoded in them.
I still disagree that writing a portion of a system allows the writer to claim that they were the creator of the larger, pre-existing system that there piece is being used in. Even if we are talking about a piece of software that is actually public domain with a completely open copyright, I could not consider myself as having created a piece of software after I download it and add a few new components. I may have created a collection of pre-existing software or pre-existing software and my own contributions, but I cannot claim that I am the creator of the original software.
I do not disgaree with your last statements. I use FOSS software and libraries and I also use proprietary software. Where I disagree the most is with Stallmans comments that the original poster was disagreeing with:
This and comments similar to this are part of the issue I have. I have software problems that aren't even close to being solved by available FOSS software. heck, I have problems that the available proprietary software is about 4-5 years behind on. Should I sink $200k-$500k into custom development (in either cash or time) of manufacturing software that I need or should I depend on cheaper, proprietary solutions that can scale to new devices and new sizes for a mch lower price tag? I choose to go with proprietary software, but according to Stallman I should have spent the money/time or consider it a job not worth doing.
Sorry for the long and somewhat general example, but wanted to clear up what I was supporting from the original parent poster.
yeah, sorry, that wasn't actually a commentary on your post's meaning, I was just pointing out that the "shortage of unwanted kids" struck me as a funny phrase :)
Do you work in the real world? IS/IT Developers develop (or find and configure) software for their customers (ie, other people who work for the company). Developers at software houses are even more obvious. Network admins, systems admins, dba's, etc all generally have deadlines to produce the goods to their 'customers', whether they be actual customers, coworkers, or other departments in their company/organization. Tech support does it. Hell, I did it in a car for five years delivering pizza.
The word "Customer" covers a lot more than it's literal definition these days. I may not care for that definition much, but I do have to admit that it exists and is in popular use.
Care to provide numbers? In House development is generally customer driven, for sale software development is customer driven, etc. Many open source projects may be just a guy/gal with no deadlines and no expectations coding a project, but the big ones have release deadlines, user groups, bugs to fix, etc. those users are just as much customers as the people in the marketing department that have decided they need a brand new website.
I once built a driver, therefore I will take take credit for having written Windows. The point being that writing a portion of something used in a system does not make a developer or organization the maker of said system. The word for that is contributor. Or developer of a seperate piece of software.
It's unfortunate, but there is no argument here to pick apart. Rather than supply an actual argument you fall back on poking the finger at Microsoft. Good job. And note that before you point your finger at my last point, the comment I made was in direct support of an argument, not a standalone random statement.
---
I think the original author did have some valid points and in a couple cases took some arguments a little far. I agree with several other posters that have pointed out that Linux would not be what it is today without GNU tools/software. But I also agree with the original poster that most of what I have read on Stallman gives me the feeling of an academic. I still recall randomly running into a mailing list for a project he was working on. For the hell of it I started reading. He spent days trying to hold back a pre-release beta because there were still outstanding bugs. The purpose of that particular pre-release beta was to gather more bug information and get real users using it to gather that information. Instead he wanted to wait until there were no (some class) bugs in it before doing the pre-release. Most of his responses came off as childish.
I support the idea of open source and if I had 4-6 more hours in the day I would probably contribute more, as it is I have only made modifications to existing projects rather than take on the responsibility of being a team member or such for one. But I have to say I don't actually like RMS. I find him to be somewhere between a whiny kid and a wild-eyed, "The End Is Nigh" madman. Someone that inspires the fans (and fanatics) that already exist, but doesn't provide the non-fans with a lot of inspiration beyond, maybe, sidling away a little to give him space, just in case