We keep this week’s podcast simple. Just two topics: Mono (finally some movement from Microsoft!) and Google’s vapid Chrome OS announcement. However, to top this up with a full bodied flavour, we cover your feedback on last week’s show.
Here’s how the audio file breaks down:
| 0:00:30 | Intro / Reader comments |
|---|---|
| 0:31:08 | “Mono†|
| 0:41:42 | “Chrome OS†|
| 0:59:50 | Meta |
| 1:03:09 | (Total Time) |
Download .mp3 |
Subscribe in iTunes |
Subscribe RSS
The intro / intermission and extro music is a Commodore 64 remix “Turrican 2 – The Final Fight†by Daree Rock.
We genuinely hope that you enjoy the show, and that we’ve managed to bring up original points in our discussion. Do follow up what you picked up on in your comments!
We are always open to your feedback. Please either leave your comments on the site, or send us an email to [email protected].
This one was good. I will listen to it again once Chrome OS ships.
I think Google will use much more of the current Linux stuff than you think (window system based on OpenGL, alsa for audio etc.) And they will certainly use O3D and Native Client.
If you post in Ogg Vorbis as well, I can actually listen to it directly with Firefox 3.5 in Fedora 11. Even on a Fedora Live CD. That is convenient.
A bit unfair this is being modded down, we did discuss this in the show.
Supporting open formats is always good especially if easy possible, elevates the much needed awareness too…
And thanks for the extra 10 kbit in mp3, sounds definitely better.
Just to let the crew know, I wasn’t referring to the difficult sane frontends, but rather the drivers themselves. Try to find a consumer scanner, still in production, that is supported well in sane. NOte that “still in production” is the key here. It’s not hard to find a scanner that will work, but it is hard to find one that will work that is still produced as by the time they are supported in sane that model has usually gone out of production.
Unfortunately most consumer scanners are now being bolted onto a printer and sold as a device that ‘does it all’. The problem is that in many cases the company needs to provide a low level device driver and thanks to the lack of either a stable user space or kernel driver API – if one were to go for it, one would be facing a constant moving target. Most hardware companies I see want to write a driver and forget about it for 3-6 months – with Linux they need to have engineers paying constant attention which sucks up much needed resources when it comes to supporting future products in the pipe line.
Oh, how well I know that, and those bolted-on all-in-ones are going downhill fast for quality. Still, sane really isn’t a moving target, and most scanners don’t need a kernel driver just a sane backend. It would actually be one of the few areas where maintaining a Linux driver wouldn’t be so demanding. Some companies, like Epson, are doing this but their drivers are binary blobs, and I don’t want that–not for philosophical reasons, but because those blobs are 32-bit only and I run a 64-bit system so they do not work properly.
It looks more and more that if I want quality in a flatbed I’ll have to either pay for a professional scanner that can cost a thousand bucks or go back to an old model anyway. They just don’t seem to make ’em like they used to.
My Canon Pixma MP160 (scanner/copier/printer) is working great on Kubuntu. I don’t know if it is still in production but i assume that some upgraded version which is on the market today would work also.
And if someone don’t like Xsane he should try Skanlite. It’s a great app for scanning.
That’s just it, most of the “upgraded versions” of such devices are using different controllers that aren’t supported, and Canon is one of the worst when it comes to this.
The all-in-ones seem to have better support than standard flatbed scanners these days, but I don’t want an all-in-one for two reasons: first, I’ve already got a printer and second, the scanners on these devices are often using CIS-based technology which does not produce the higher quality scans I need to handle some of my OCR tasks. Ah well, it seems to get a high quality flatbed scanner these days I have to go back to the older models anyway. Might as well pick up an Epson Perfection 1650, those things are tanks and compatible with basically every os out there.
first of all thanks for the new podcast, always a pleasure to listen to those
You’re always talking about how hard it is to install drivers on windows, you are always referring to XP, an 8yo os. Since vista (7 has increased its driver base) all the drivers are installed and updated without any action on the part of the user (if one exist), printer too. That completely mess up the whole printer driver.
The problem with Windows Vista is that they kept the old cruft for backwards compatibility. They developed a wonderful replacement for the GDI+ based printing in Windows XP but they tried to graft on compatibility with old printer drivers – the net result? rather than having a nice black and white, works and doesn’t work scenario, you get this horrible merky grey area where it is Russian roulette whether it works or not.
Then you have the hardware vendors who make the least amount of effort in moving their drivers to the new printing infrastructure; removing all backwards compatibility would have forced them to write new drivers rather than retrofitting using a frankenstein monstrosity that is enough to get it working but still relies on the old infrastructure.
Did you not know that CUPS has worked in OS X since 2002 and has been included in the operating system since 10.4, I believe. So, you might not be able to say that using a printer is better supported on Linux than on a Mac… especially considering that Apple own CUPS!
🙂
Michael R. Sweet, its developer works for Apple.
But good job guys. Enjoying the podcasts.
Would like to hear your comments on what would be in a dream OS that could put Windows or the Mac OS in the shade. You’re up on the latest stuff, but from a readers perspective, I find I can miss some of what’s going on, so I’d like to hear your opinions as that’s the fun stuff for me. Dreaming of something better… What features would make a difference? Looking forward to see what you thought of the Amiga Thom and how that’s different, as you mentioned on the podcast.
Cheers!
Printer support has nothing to do with CUPS – the printers supported are provided through drivers from both Gütenprint and drivers from the printer companies themselves. CUPS provides no printer support itself, CUPS is only an implementation of IPP (Internet Printing Protocol).
CUPS includes drivers for Dymo, Epson, HP, Intellibar, Okidata, and Zebra printers.
Yeah, I’d forgotten about GIMP-print and they work in OS X too…
I agree it would be great to have an edition on “what are the features a dream OS should/would have”.
For example, I remember you were talking about the time slider of (open)solaris a while ago, that’s for me one of the many things, support for many FileSystem, etc. It’s the kind of topic people who tried many OS could join you to discuss what they tried out (especially features on the server side that could be translated on the client side like shadow copy that you talked at the same time of time slider)
To me it’s live mesh all over, files I need to sync, all my application’s settings/favorites/contacts/etc
We would always end up coming back to BeOS. All the basic principles were there, it just didn’t get the opportunity to mature those ideas parallel to what we have now.
Hi Kroc!
If BeOS had all the underpinnings to make a perfect operating system, it’s interesting that Apple wouldn’t pay the $200 million asking price, but would pay over double that for NeXTstep. If BeOS had it all, does that imply that Haiku, the free rewrite of the OS, is going to develop into the perfect operating system? I hate to ask, but why is that then? I come to learn from those who know!
So, in your eyes, discussion of future developments of OS interfaces, spoken, gestural, 3d, augmented reality, security, software distribution, packages, system architecture, distributed, web, always on, instant on, integrated, embedded and converged in products will always bring us back to BeOS… OK, is Haiku all our futures?
I would guess everyone who comes to OSnews is dreaming of changes to make life better, what will those changes be? If Linux is now so patched and layered that it’s now unstable or, at least, not optimal, what comes next?
I’d like people’s opinions on why no one is developing an OS on Apple’s Darwin kernel, if it’s good enough for what is generally thought of as the best OS out there now. What is the most rationalised kernel to build on? Why did Apple choose their direction and what would they choose now or would they start from scratch? Why did google choose Linux for Chrome? Where should Windows go from here? Start afresh? With what?
I’m sure that everyone has some part of the answers. I know Thom believes there are better ways as he’s written of his own ideas. Would love to hear more opinions.
I’m not sure what you’re asking really. Haiku is a reimplementation of a 2001 OS, it is not a 2009 OS designed for 2009. Compatibility first.
Had BeOS been the No.1 OS, or at least in a decent situation like OS X, then the period of 2001–2009 would have seen steady improvement of BeOS into what I would deem the best OS ever.
As it stands Haiku still hasn’t got on its feet and certainly will never be mainstream.
I suppose what I’m asking is, if BeOS was so good in its time, if it was the best foundation for future development then, would it still be now? If not, what would be the best foundation to develop on now? If it was the best foundation of ideas, then, why wouldn’t there be more push with Haiku or something similar?
I’m interested to learn about what makes the best architecture and why? There are going to be lots of opinions. Thom has just asked, what features are lacking in your OS? I think most people would agree on a core of features they would like to see, but how do you get there? What influence does the chosen kernel have on creating a foundation for feature completeness. Perhaps it’s only the developers working on the kernels who really start to see the limitations of that kernel, and after such investment to develop the thing, it’s impossible to make major architectural changes. For example, I read rumours that the Mach kernel in Darwin is a limitation, but perhaps Apple are stuck with it.
You might not have those answers Kroc, and perhaps others at OSNews may not, but the reason I come to your site is to find those answers, to raise my awareness, to anticipate where the industry is going so that I might support those who seem to know those answers and are trying to bring about something better. They are not easy questions to answer, for me, operating systems are so complex, but I’m sure that the readers, between them, know those answers… which is why I’d love to hear invited opinion, open opinion and discussion on such matters.
Would very much like to see a series of articles on the subject of how to get to where we want to be. Perhaps invited opinion pieces by people in the industry or through a call for voluntary submissions. Perhaps just posing provocative questions for comment as Thom has just done.
Thanks for all the work you must all put in to bring us the news, research and write your pieces. You’ve made OSNews my favourite site to visit.
Apple bought NEXT because it got them Steve.
I myself wouldn’t mind being part of the podcast, but I would think most of my comments would be, when Thom interrupts Kroc again and again (with the same point spoken faster), “For God’s sake, let the man speak!”
No it’s fine, if anything I feel like I interrupt him and he never interrupts me. We get where we want to go in the end anyway. If you do want to be on the show just pop us an e-mail.
it’s true that you’re always refering to beOS, but an OS who stopped in 2001, before the huge impact the Internet had on our lives cannot possibly contain everything we might wish for now …
. Although I like the reference to the OS even if I never even tried it and haven’t read about it yet ^^