The biggest new feature in both Ozone and v.Next is the .NET Compact Framework in ROM. This is a run-time layer that isolates applications from the operating system, managing memory and enforcing security. It should prevent buggy apps from crashing the device. Read the article at NewsFactor. Head over at NewMobileComputing for more mobile technology news. In another market arena, the market share for Windows in Network Attached Storage devices rose 8 percent in the first quarter to 41 percent.
So would some good old fashioned Quality Control…
Uhh, what software dosn’t have bugs ?
Apart from a few ‘Hello World’ apps, none. But my argument still stands. Some software is *much* better than others. Try harder next time.
You are not alluding to UNIX, or are you?
Hmmm. I wonder if Windows-NAS devices have internet exploder on them.
Prolly blowzilla/slowzilla.
Of course, it’d have to be Mozilla, because it’s integrated into the OS.
Thanks!
.Net bugs, in ROM. Wait for that first bug — you know it’s there, Microsoft wrote this software — and there’s no way to patch it.
“You are not alluding to UNIX, or are you?”
Parts of it. Simple fact is that smaller programs are less likely to be buggy. Examples being ‘cat’, or ‘cp’. Those things are about as perfect as software can get.
James, I dare you to find a crash-inducing buffer overflow in one of those.
I’m really tired of hearing all these anti-microsoft comments here. Especially the one’s making fun of .NET, which is really a very cool architecture.
I can’t possibly describe enough detail here to make anyone understand why it’s interesting though. So instead I’ll just post a link to this excellent overview by Arstechnica.
http://www.arstechnica.com/paedia/n/net/net-1.html
Please note especially the section on saftey.
You may hate Microsoft. They may be a monopoly. But .NET is good and open. It’s based on open standards, and is itself an open standard (EMCA, it has it now, although at the time of the review it was still pending).
RE: kingston
UNIX software may have very few buffer overlows, but no matter how bad a .NET program is coded, it *can’t* have a buffer overflow. The system simply doesn’t allow it. It is managed code, similar to Java. You can’t have memory leaks, you can’t have buffer overlows, you can’t cast an object to the wrong time. It’s not possible. The runtime and compile time systems prevent it 100%.
What is the definition by managed code?
“But my argument still stands. Some software is *much* better than others. ”
Being *better* doesn’t mean *bug free*. That said, having a system that doesn’t crash when a software (even your better one) trigger a bug, it’s a HUGE plus.
Try harder next time.
“.Net bugs, in ROM. Wait for that first bug — you know it’s there, Microsoft wrote this software — and there’s no way to patch it.”
When you finally get around to learning about the mobile industry you’ll find that Microsoft is well ahead of others in technology to provide updates to devices. It’s very easy to patch something in ROM with a RAM installed update and the infrastructure is already going to be there for ISVs to use it themselves. In fact, the updates are like Windows Update (which is very effective) and can even be automatically enforced by network operators so cases like SQL Slammer (where the patch was out 6 months ahead of time but people didn’t install it) won’t be able to happen.
You can research this or you could continue to FUD and make stuff up–either way the devices are already quickly becoming the best out there, although there aren’t many yet. I’ve been working with the .Net CF for over a year now and will never use J2ME again because the limitations and fragmentation make life impossible as a developer, even if it is on millions of devices.
You can also flash the ROM with updated code.
PocketPCs, for example, have had several vendor-supplied ROM updates to add features and fix bugs. Most Pocket PC 2002 devices are flash upgradable to Windows Mobile 2003.
<But .NET is good and open. It’s based on open standards, and is itself an open standard (EMCA, it has it now, although at the time of the review it was still pending). >
Do you have a link ?
The only thing I worry about is the usual MS bloat. Will .NET Compact Framework become .NET Framework next year? I don’t want to have to start replacing PDAs every couple of years.
Offhand ..
http://msdn.microsoft.com/net/ecma/ would be a good start.
Boo, hoo. Microsoft is a cuddly and likable as a tarantula. We’re tired of <a href=”http://dictionary.reference.com/search?q=thrall“>thralls</a&g… rushing to the defense of an extremely rich and criminal software company, regardless of what Microsoft does.
.NET is *not* open. C# has been submitted to ECMA for standardization. Windows.Forms, for instance, has not.
How might one flash a Read Only Memory? Perhaps you’re talking about EEPROM?
It seems that most of the MS-based have client maximums, usually 25. Since I don’t think that the OS / hardware is incapable of only 25 connections, so this artificial maximum clients forces me to look to make sure that any NAS appliance is not running MS. Sometimes the client information is buried in the specs.
http://h18007.www1.hp.com/support/files/handheldiPAQ/us/download/18…
Perhaps they practice black magic ?
Just as I expected…instead of commenting on the actual article, we get the same old, tired, uninformed, rhetoric:
MS is evil, .NET is evil, blah blah blah
Okay, some people hate .NET because it comes from Microsoft, fine.
I’d love to see someone give a good argument against based on technical reasons.
Here are some non-MS links showing standards.
http://www.ecma-international.org/publications/standards/stnindex.h…
http://news.zdnet.co.uk/story/0,,t269-s2132925,00.html
C# and the CLI both *have* ECMA *and* ISO certification. They are as open as you can get. (Note we are not talking about implementation here).
The CLI and C# are really the most interesting parts of the .NET platform. They are the critical infrastructure and anyone who doesn’t use Windows wouldn’t care about anything else.
re: anonymous
Some areas like Windows forms are not open. But those are designed to be used on Windows alone. Something about your venom filled comment tells me you don’t use Windows. So why should you care if Forms are open or not. Forms are not usefull on other platforms anyways.
That said, there is work to implement the Forms api over GTK I believe. I defend technical excellence. If you want to bash Microsoft, do it with at least SOME intelligence please. If you want to say they are a Monopoly and are evil, fine. The fact remains, .NET is a great system. And since you are nitpicking, the *important* and relevant parts are open standards.
Re: managed code
Managed code usually reffers to code which is *managed* by a runtime system. The runtime will do things like check for buffer overflows, incorrect casts, and also managed memory deallocation (garbage collection). Java is also a managed environment, perhaps even more so than C#. A lot of people find Java a little too managed though. It’s difficult to interface with outside libraries using Java, although there are tools to make this easier. C# and the CLI were designed specifically to expose parts of the operating system. While this hurts portability somewhat, it allows the whole system to be more flexible. It’s all about tradeoffs, Java was the first large scale experiment with managed code. C# and the CLI were able to learn from Java’s mistakes and have corrected most of them.
“We’re tired of thralls rushing to the defense of an extremely rich and criminal software company, regardless of what Microsoft does.”
I’m proud to defend a company that give me as great products as XP, Office & Visual Studio. The day I can’t say that about those, then I’ll stop defending MS. Until then, they are a software company with an awesome product line that I enjoy most than any other softwares to use and work with.
“How might one flash a Read Only Memory? Perhaps you’re talking about EEPROM? ”
And why not ? Flashable ROMs (EEPROM & al) in 2003 are so cheap to produce, that it’s unlikely Microsoft will go the road of closed, fixed ROM. Yet, if I’m wrong, then I agree it’s a crappy plan. But I’ll like to get more information on that before.
He was nitpicking, Pessimist.
My comments clearly implied EEPROM. The point was that MS’ mobile hardware partners have been using EEPROMs for some time now and issuing upgrades and feature additions that way, as well as over the network in the case of Smartphone, as another person mentioned. Thus, the poster talking about unpatchable code in ROM has no reason for concern beyond the level of support offered by the hardware vendor (HP, Dell, T-Mobile, etc.).
“Parts of it. Simple fact is that smaller programs are less likely to be buggy. Examples being ‘cat’, or ‘cp’. Those things are about as perfect as software can get. ”
That’s a good comment : cp and cat are typically the kind of app that someone need on his mobile.
Congratulations for one of the dumber comment I have ever seen here.
With the recent talk about winforms being a windows only thing is just nonsense. It has been clearly shown that a compatible layer can be built between winforms and x11. Portable.Net has been doing this for some time now. It even works on handhelds that run tinyX, like Familiar Linux. I like the idea of building my gui and being able to run it on any device without recompiling or needing a special o/s =)
I’m sure it works as well as Windows 95, but it doesn’t even provide enough basic security just to surf the web.
I have a few problems with this news report. 1. 41 percent of what, total market share? And this is news?
News would be them getting convicted for numerous violations of the Sherman and Clayton acts. Are we really concerned with how well it would perform, now?
Not really, as it takes too long for them to get anything right. Time is money. My money.
No thanks :-}
Most Pocket PC 2002 devices are flash upgradable to Windows Mobile 2003
Haha, like the 1 year old toshiba e740 for example