|
|
| Author |
Message |
CC
none
Joined: 15 Jul 2003
Posts: 6
|
 Posted: Dec 21, 2003 7:53am Post subject: |
 |
|
| I must commend each and everyone who has contributed to this thread. A very interesting read and has given myself quite a bit to think about. Sometimes the IRC community in general surprises me with how much introspection they do to better themselves. Things have changed from the past, but instead of us wishing that it would revert to how it was, let's embrace the future and make sure that IRC is not eradicated in the process. |
|
| Back to top |
|
 |
ed
SearchIRC Staff
Joined: 25 May 2003
Posts: 366
Location: Baton Rouge, LA
|
| Posted: Dec 21, 2003 10:32am Post subject: |
|
|
| It could actually be accomplished easily. As long as your services use MySQL, all someone would have to do is write a WAP interface. |
|
| Back to top |
|
|
Wiggle
Newbie
Joined: 01 Nov 2003
Posts: 81
|
| Posted: Dec 21, 2003 11:57am Post subject: |
|
|
True, ed. I should think before I post things  |
|
| Back to top |
|
|
uchat
Idler
Joined: 17 Mar 2004
Posts: 335
|
| Posted: Mar 17, 2004 8:58pm Post subject: |
|
|
why not memo to phone?  .. there is memo to email (which can go to phone) and web based memo scripts. |
|
| Back to top |
|
|
Itsy
none
Joined: 10 Mar 2004
Posts: 5
|
| Posted: Mar 26, 2004 10:13pm Post subject: Observation on basics |
|
|
It seems from watching this thread that there are still folks out there that think there is some advantage or eliteness to being a coder. I can write code, but i'm not remotely interested in writing my own browser or word processor, or text editor. Yet there are dozens and dozens of such things. How many versions of wordpad does the world need?? (possibly less than 1, but I digress..)
IRC is now a 'commodity' .. like any other product. Real people can grab and go (for better or worse) and establish a n online community of interest -- which is what IRC is actually all about anyway. "community" and "interest" .. without both, the net probably wont grow and certainly wont contribute much to irc in general. But, they're allowed to. I get a little suspicious when someone wonders if a new net is 'good' for anything or improves the world. Most things human don't improve the general state of the universe, and IRC fits in that category.
Just as the internet took communication of of the hands of publishers, the 'commoditization' of IRC software and connectivity has taken it out of the hands of programmers, geeks, and other founding fathers of cyberland and put it into the hands of the actual people who want to use them. VERY VERY few coders want to BE in an IRC network.
Anyway, my 4c on the subject. The direction of IRC is going to be towared enabling joe and josephine to make a contact point and community without needing to learn programming, or even much of the internet.
Newsletter publishers dont usualy know a whole lot about the ins and outs of publishing and the printing biz. THey dont need to. At one time, they did, because it was horribly techinical and complex just to print a single issue. Now anybody with a printer is a publisher. This is the direction IRC is headed and if we're all smart we'll get on board with it and figurre out how we can support the new paradigm. |
|
| Back to top |
|
|
codemastr
Idler
Joined: 05 Feb 2004
Posts: 345
|
| Posted: Mar 26, 2004 11:19pm Post subject: |
|
|
I just wanted to respond to the comments here, as an IRC enthusiast, admin, and programmer.
| Quote: | | Why, I remember a time when every network had a programmer who developed their own stuff, and put an awful lot of time and effort into that sort of thing |
Well, I'll admit, I haven't been using IRC since it was invented, but I have been using IRC for about 6 years now. In that time, 90% of the networks I chatted on did not have their own programmers. Yes, the DALnet's, EFnet's, etc. did, but even back then, the little nets didn't. The only way I can think what you're saying is true is if you're going very far back. For example, in the late 80's, early 90's of course most networks had programmers, but there were maybe 5-10 networks in existence. Whereas today, there are thousands. So it's not really easy to make that comparison.
| Quote: | | Do they know C programming now? Maybe not. Can they learn it? Sure. |
Could they? Of course they could. But let me use myself as an example. I began programming when I was 8 years old. At current, I'm almost 20 now. So I've been doing programming for more than half my life. On average, I spend anywhere from 5-10 hours a day programming (both in school and for my own projects at home). I've bought several thousands of dollars worth of books, I've spent tens of thousands of dollars in college to take classes. After all of that, I'd consider myself a "novice" C programmer. Yes that's right, with 12 years of programming experience, 5 of which were with C, I'd consider myself a novice. There are far more things I don't know how to do than I can do.
So now the question is, do you want to make that commitment? Sure, if you were like me, and when you were 8 years old when you saw your first BASIC program, you knew computers were what you wanted to make your career, you might want to put in the dedication necessary to become a good programmer. However, I doubt most people know at 8 years old what they want to do with the rest of their lives.
This is exactly why things like Unreal and Anope exist. We realized, not many networks have the resources necessary to either persuade programmers to join (after all, how enticing is the promise of an o-line on a network with 20 users?) nor could they afford to hire one (A good programmer can cost anywhere from $50-$150 an hour). So what we did was we created an IRCd that could be easily customized to provide a unique experience for each network using it. Additionally, we also provide custom programming services (for a fee) so that people who want customizations can have it done without having to pay the high prices it would cost to hire a programmer. I know I've worked on projects for many networks that vastly customize their IRCd in different ways.
So basically, could everyone become a programmer? Of course, but "should" everyone become a programmer? Probably not. In my mind, being a programmer takes a special kind of person. It takes someone who isn't afraid to spend weeks to solve a problem that might be very small (I once spent 5 hours, every day, for 6 months, searching for a bug in Unreal that turned out to be a misplaced ";"). You also have to enjoy problem solving, I've sometimes written 30-40 page documents describing what I want to do and how I'm going to do it. Then, I think about it some more and realize my entire logic was flawed and have to write that entire document over again. Being a programmer and software designer is not an easy task, it is not something you can learn overnight. It is something that takes years to get good at. So unless you really want to put the time into learning how to do it, and how to do it well, it's best to leave it to the professionals. |
|
| Back to top |
|
|
magpie
Idler
Joined: 18 Jan 2004
Posts: 453
Location: Essex, UK
|
| Posted: Mar 27, 2004 6:07am Post subject: |
|
|
| codemastr wrote: | I just wanted to respond to the comments here, as an IRC enthusiast, admin, and programmer.
| Quote: | | Why, I remember a time when every network had a programmer who developed their own stuff, and put an awful lot of time and effort into that sort of thing |
Well, I'll admit, I haven't been using IRC since it was invented, but I have been using IRC for about 6 years now. In that time, 90% of the networks I chatted on did not have their own programmers. Yes, the DALnet's, EFnet's, etc. did, but even back then, the little nets didn't. The only way I can think what you're saying is true is if you're going very far back. For example, in the late 80's, early 90's of course most networks had programmers, but there were maybe 5-10 networks in existence. Whereas today, there are thousands. So it's not really easy to make that comparison. |
I've seen quite a few little networks started up by programmers who were bored, or wanted somewhere to test their own services: eventually establishing a user base. This even happens when programmers from larger nets get tired, for various reasons, and leave to found their own network.
| codemastr wrote: | | Quote: | | Do they know C programming now? Maybe not. Can they learn it? Sure. |
Could they? Of course they could. But let me use myself as an example. I began programming when I was 8 years old. At current, I'm almost 20 now. So I've been doing programming for more than half my life. On average, I spend anywhere from 5-10 hours a day programming (both in school and for my own projects at home). I've bought several thousands of dollars worth of books, I've spent tens of thousands of dollars in college to take classes. After all of that, I'd consider myself a "novice" C programmer. Yes that's right, with 12 years of programming experience, 5 of which were with C, I'd consider myself a novice. There are far more things I don't know how to do than I can do. |
C obviously isn't right for you. It's generally the larger networks that are limited to C because they _need_ the speed. It is also substantially more difficult to write 'secure' code in C.
| codemastr wrote: | | ...(after all, how enticing is the promise of an o-line on a network with 20 users?) nor could they afford to hire one... |
I've always believed using an O: line to entice people onto your network is a bad idea. Sure, if the person is suitable then oper them, but it should never be expected imo.
Programmers are not the only part of an IRC network anyway. Larger nets often divide staff into appropriate groups (I can't comment about smaller nets), such as development teams, 'operations' teams and so on. Arguably, the most important part of an IRC network is its support staff. This is what I believe can truly set one network apart from another. |
|
| Back to top |
|
|
codemastr
Idler
Joined: 05 Feb 2004
Posts: 345
|
| Posted: Mar 27, 2004 11:56am Post subject: |
|
|
| Quote: | | I've seen quite a few little networks started up by programmers who were bored, or wanted somewhere to test their own services: eventually establishing a user base. This even happens when programmers from larger nets get tired, for various reasons, and leave to found their own network. |
Operative word is "few" even if you've seen 50, that's not even 1% of the IRC networks out there. So yeah, it does happen, I mean that's how it happened with me, but, that's not what happens on most networks. And, as was said about these "cookie cutter" networks, if you go and look at the IRC software list here, you'll see the majority of servers are running Unreal, Hybrid, Bahamut, or Ultimate. You'll see only a handful running their own software.
| Quote: | | C obviously isn't right for you. It's generally the larger networks that are limited to C because they _need_ the speed. It is also substantially more difficult to write 'secure' code in C. |
C is just perfect for me, thank you. I can write programs in 9 different languages ranging from Assembler to PHP, C is the language I've chosen. Also, software design has NOTHING to do with the language chosen. Programming is the easy part, anyone can learn that, as I said, it's the designing that is what is hard and takes years to master. As far as programming goes, if there is something you don't know, you go to Google and you find out, when it comes to designing, you can't do that.
In any case, your whole statement is wrong. C needed for ALL networks. Why? Because 99% of IRCds out there are based on C. So unless they're going to hire a coder to write an IRCd from scratch, they need someone who knows C. Furthermore, Unreal is used by several large networks, therefore they do need speed so even if I had the option to write it in some other language, I'd still choose C. And lastly, it's only hard to write secure code if you're not good at it. Writing secure code is not very hard at all if you know what you're doing. Of course, every now and then a bug slips by, but you could be the best programmer in the world and that would still happen. Once you know the common mistakes (buffer overrun, memory leaks, referencing null, etc.) they are pretty easy to prevent.
In any case, what I meant by "novice" is not writing the code, that's super easy. It's coming up with the ideas and then finding a way to translate those ideas into C. Remember, C does virtually nothing for you. It has very basic library routines, almost everything is up to the programmer. Therefore, taking an idea and operationalizing it can be very difficult in C. That is what I was referring to, the programming itself is easy. That's why software designers sometimes make 10 times what a programmer makes. There is a difference between a designer and a "code monkey." I specifically mentioned a novice at C rather than a novice software designer simply because most of the other languages make it much easier for you to design things. You want to download a website in PHP, ok just use the fopen() function. You want to do it in Java? Ok, just use the URL, URLConnection, etc. classes. Want to do it in C? Ok, go make a socket connection, learn the HTTP protocol, and handle it all yourself. Of course external libraries exist to do this, but they aren't part of the C language. Code which in most languages would be a few lines, in C is hundreds, if not thousands. It requires you to be able to come up with a design and find a way to code it all by yourself.
| Quote: | | I've always believed using an O: line to entice people onto your network is a bad idea. Sure, if the person is suitable then oper them, but it should never be expected imo. |
I agree, however, small networks don't generally have much else to offer to programmers.
| Quote: | | Programmers are not the only part of an IRC network anyway. Larger nets often divide staff into appropriate groups (I can't comment about smaller nets), such as development teams, 'operations' teams and so on. Arguably, the most important part of an IRC network is its support staff. This is what I believe can truly set one network apart from another. |
Indeed, but, it doesn't take very long to learn how to route servers, how to manage klines, etc. It does take a long time to learn how to program. |
|
| Back to top |
|
|
Guest
|
| Posted: Mar 27, 2004 12:30pm Post subject: |
|
|
| codemastr wrote: |
In any case, your whole statement is wrong. C needed for ALL networks. Why? Because 99% of IRCds out there are based on C. So unless they're going to hire a coder to write an IRCd from scratch, they need someone who knows C. Furthermore, Unreal is used by several large networks, therefore they do need speed so even if I had the option to write it in some other language, I'd still choose C. And lastly, it's only hard to write secure code if you're not good at it. Writing secure code is not very hard at all if you know what you're doing. Of course, every now and then a bug slips by, but you could be the best programmer in the world and that would still happen. Once you know the common mistakes (buffer overrun, memory leaks, referencing null, etc.) they are pretty easy to prevent.
|
It would be easy to write an IRCd for emacs  |
|
| Back to top |
|
|
magpie
Idler
Joined: 18 Jan 2004
Posts: 453
Location: Essex, UK
|
| Posted: Mar 27, 2004 3:18pm Post subject: |
|
|
| codemastr wrote: | | In any case, your whole statement is wrong. C needed for ALL networks. Why? Because 99% of IRCds out there are based on C. So unless they're going to hire a coder to write an IRCd from scratch, they need someone who knows C. |
Who cares what language the IRCd is written in? You said yourself most small networks don't use their own, rather somebody else's. I should imagine most small networks don't even need to touch the IRCd source code, they simply run the software.
Also, services don't have to be written in C, regardless of the language the IRCd is written in. I could write some services in Delphi if I wanted to (assuming I actually knew the language) and they'd work with the IRCd I'd chosen (assuming I stuck to the correct protocol).
| codemastr wrote: | | Furthermore, Unreal is used by several large networks, therefore they do need speed so even if I had the option to write it in some other language, I'd still choose C. |
Uh, didn't I say that the big nets _need_ C for speed? Or have I misunderstood your point?
| codemastr wrote: | | And lastly, it's only hard to write secure code if you're not good at it. Writing secure code is not very hard at all if you know what you're doing. Of course, every now and then a bug slips by, but you could be the best programmer in the world and that would still happen. Once you know the common mistakes (buffer overrun, memory leaks, referencing null, etc.) they are pretty easy to prevent. |
Yes, that's why there are so many Windows exploits out there; the recent spate of vulnerabilities in the Linux 2.4 kernel also spring to mind (don't ask me to quote specifics, I just remember reading about them in general).
The fact is, with any large project you're going to get bugs. New and inexperienced programmers are pretty obviously going to introduce more than the older, more experienced ones.
| codemastr wrote: | | In any case, what I meant by "novice" is not writing the code, that's super easy. It's coming up with the ideas and then finding a way to translate those ideas into C. Remember, C does virtually nothing for you. It has very basic library routines, almost everything is up to the programmer. |
I'm aware of what libraries are and do. Perhaps I was mislead by your original comment about spending tens of thousands of dollars on classes and books, but still considering yourself to be novice.
| codemastr wrote: | | Quote: | | I've always believed using an O: line to entice people onto your network is a bad idea. Sure, if the person is suitable then oper them, but it should never be expected imo. |
I agree, however, small networks don't generally have much else to offer to programmers. |
I should imagine there's a distinct lack of politics. You could argue that's something very tempting they have to offer. :)
| codemastr wrote: | | Quote: | | Programmers are not the only part of an IRC network anyway. Larger nets often divide staff into appropriate groups (I can't comment about smaller nets), such as development teams, 'operations' teams and so on. Arguably, the most important part of an IRC network is its support staff. This is what I believe can truly set one network apart from another. |
Indeed, but, it doesn't take very long to learn how to route servers, how to manage klines, etc. It does take a long time to learn how to program. |
To program, or to plan? :) |
|
| Back to top |
|
|
codemastr
Idler
Joined: 05 Feb 2004
Posts: 345
|
| Posted: Mar 27, 2004 7:28pm Post subject: |
|
|
| Quote: | | I should imagine there's a distinct lack of politics. You could argue that's something very tempting they have to offer. |
Indeed, but as this thread started out, once you have a unique network, then you might begin to grow. Enter IRC politics, exit IRC programmers. Meaning if unique features do indeed bring more users, you can expect as a network grows, more political nonsense will enter in. That seems to have been what lead to the downfall of just about every network I've ever been on the staff of. People start arguing, people start abusing power, next thing you know, it's not a fun place to be anymore. |
|
| Back to top |
|
|
|
|
| |
Register
Log in
