Whither 802.11a in Linux? 23
Revar asks: "My local net admin recently installed an 802.11a 54Mbps wireless network. Under Windows, the speed is great and at 5Ghz, it has much less interference then the 2.4Ghz 802.11b wireless. The problem is, I cannot seem to find any 802.11a PCMCIA cards that have Linux or MacOS X support. Are there any, or is no-one actually working on this?" Whenever new hardware is released patience, when wating for Linux support, is a necessary virtue.
Re:Virtue (Score:4, Insightful)
who modded this up?
Easy to write device drivers! Hmmm, for some prehaps but for most of us it's not. Next time you need a new piece for your car I hope the mechaninc points you to a lathe.
Re:Virtue (Score:2)
Hmm...
reminds me of something else entirely...
Re:Virtue (Score:2)
No, its not necessary at all. A much more neccessary (and more easy to obtain) virtue would be the motivation, and consideration for the open source movement that bought you linux in the first place, to attempt to actually hack up a quick driver yourself...
Oh, yes, of course - because every Linux user out there knows how to program device drivers. How silly of them to think that the Linux movement actually wants people other than programmers to use the operating system.
Re:Virtue (Score:3, Insightful)
It depends on what you mean by "the Linux movement". Sure, everybody wants lots of users. But very few people are interested in signing up to be the personal slaves of people who don't know a one from a zero and would like to stay that way.
A lot of open-source software gets written under the itch-scratching model. I want software to do X, so I write it until I'm happy, and maybe I write a little more because it's fun. Somebody else who wants X+Y writes the Y. But if no programmer wants X+Y+Z, then Z may never get done, no matter how many non-programmers want it.
This isn't meant to be exlusionary. Quite the opposite: a lot of Zs do get written for fun or out of generosity, and I've never dealt with an open-source developer who wasn't immensely welcoming towards those who wanted to learn enough so that they could contribute.
But by and large, the rule is "patches welcome". If you think "the Linux movement" should have better driver support, then you should start writing drivers or assist somebody who can.
Sure, all hardware is documented all the time... (Score:2)
d-link 650 (Score:1, Insightful)
Re:d-link 650 (Score:1)
Re:d-link 650 (Score:2, Informative)
802.11 *a* not *b* (Score:4, Informative)
The 650H (high-power 100 mW) does not work well at all - These appear to be repackaged Symbol cards - Expect frequent lockups, if it works at all. Someone was working on a driver for the Symbol Spectrum24, but work stopped.
The 650+ (Enhanced 802.11 - 22 MBps, apparently a "halfway" implementation of 802.11g) uses the TI ACX100 chipset - Also no drivers for this exist.
And for 802.11a, the DWL-A650 has no support whatsoever.
Re:802.11 *a* not *b* (Score:1)
My buddies black 650 works fine and automagically under redhat, mine seems to work under RH, but just seems to lockup after a random number of minutes, only way to get it working again is reboot. Then it works fine??? Any ideas or thoughts on what I can try???
Re:802.11 *a* not *b* (Score:2)
All 650s I've seen have a grey antenna housing, not black. It's a sort of "mushroom" shape - Wider at the tip than the base, with a single green status LED. Works BEAUTIFULLY with the linux-wlan package (I know it'll work with the Orinoco drivers, but gives some errors and in general makes me paranoid...)
The pictures of the 650H I've seen have a "tombstone"-like black antenna housing. (Which jives with the suspicion that they are repacked Symbol cards.)
Agreed, with a reservation. (Score:1)
I almost entirely agree with your sentiment, but I don't believe that software could truly be considered Free unless it is free for everyone - even rude, greedy imbeciles.
the problem is also with people like you.. (Score:1)
OK I will.
Just provide me(a person with some experince programming and some electronics) with the information on how to start writing drivers. How do drivers in linux work? How does one start probing a piece of hardware to figure out how to interact with it. How do I write the driver once I have that info? Give me an example of how someone took a new piece of hardware probed it and figured out how the driver should be written, a step-by-step example and the thought proccess involved. Not just the I probled these memory addresses and wrote this line of code to do this and this line does this...I mean the whole friggin thought proccess that one could read and use to learn to do it on other hardware.
And for that matter keep going once you start documenting, explain how to write everything for linux so people like me who only got tought MS programing in school can help The Cause(TM). [OK so I also learned shell scriptin, but thats not really programming] And while your at it GPL the documenting and post it as HTML, and link to it when you rant like above. At least with Windows if I can't find a piece of software I HAVE to have, I can write it up....and I suspect I'm not the only one in this position. Hey being real for a second, MS shitty as they are, VB is easy to use. So is their C++ in studio 6.0
One of the best
Cisco cards will have Linux and OS/X support. (Score:4, Informative)
Right now, their 1200 access-point does 802.11a, but it's primarily focused on AP-to-AP wireless backbone connectivity (the AP can hold both an 802.11a and an 802.11b radio simultaneously). When they come out with the cards later this year, they'll give you the driver support you want.
Re:Cisco cards will have OS/X support. (Score:1)
Some info on possible 802.11-a Linux support (Score:3, Informative)
The short of it is a description of 802.11-a. The main page [hp.com] contains tons of info to get you started working on your own drivers -- or there may be some gems in there too (I.e. identifying a 5 GHz card or driver that works with your system).
-Turkey