Template talk:Table mobile operating systems

From Wikipedia, the free encyclopedia

Contents

[edit] Notes about this article

Over the past decade, mobile operating systems have developed from simple notetaking and scheduling systems into powerful operating systems with features comparable to typical desktop operating systems.

This article provides a list of the popular mobile operating systems, and footnotes indicating the provision of certain features that enable powerful functionality similar to that of desktop operating systems. Given that around half of the current mobile operating systems are based on Linux, the list also indicates whether or not the mobile operating systems are Linux-based.

--InternetMeme (talk) 12:24, 20 April 2008 (UTC)

[edit] Bullet points

Hey, by your edit summary, do you mean that the order you put them in is in accordance with their market share?

Also, I read somewhere on Wikipedia that bullets are somewhat discouraged, and since there is only one list with all the OSes now without the two sub-lists, I thought the bullets didn't really add any clarity like they did before. And they also take up precious horizontal space, and squash all the items together vertically. I'd far prefer to do away with them now. What do you think? InternetMeme (talk) 11:28, 17 April 2008 (UTC)

I've removed the bullet points for the following reasons:

  • Given that there are no sub-lists, the bullet points don't aid clarity.
  • There really isn't enough horizontal room for them.
  • They make all the OS-list items squash together vertically.
  • They visually demote the OS-list items to the same level as the footnotes.

Thanks for your understanding : )

I've also spaced out the footnote indicators, as it can be very hard to distinguish the letters on a high-res CRT : ) --InternetMeme (talk) 12:22, 20 April 2008 (UTC)

[edit] Native code

I understand the concept of machine/native code. The information I'm providing with the "native code" tag is to help people who are interested in running custom native applications on their devices. Currently, Blackberry OS, Android, and LiMo Platform don't allow this. InternetMeme (talk) 15:44, 17 April 2008 (UTC)

1. I am glad you understand the concept of native code; but that doesn't make writing the wrong thing correct. All the platforms run native code; it's just that some restrict who can run it.

2. Yes, some people are interested in running native code without getting certification - basically a few hobbyists who want to run C code on a mobile. Some people are interested in running, say Java. That's a tie, and anyway, ***anyone capable of writing native code is more than capable of finding out whether a platform can run it in a few second of googling***. But even without the last, the security trumps every other issue, especially given the growing opportunities for denial of service attacks on telecoms infrastructure. Its security more than anything else that categorizes platforms; showing the absence of security as if it is a positive feature is very, very silly.

3. It's misleading and incoherent to have the categories "native code" and "certified" - certified is a form of native code.

4. If you want to make information on which platforms can run uncertified native code available to hobbyists, then put it on a webpage. Wikipedia's intent is provide info to a general audience, not act as a repository for hobbyists, whether they program in C or Java, so a large general factor - which security is - necessarily trumps a hobbyist motivation in editing and structuring information.

5. If the chart shows managed code and certified code capabilities, then it's a chart to provides security info - a symbol equates to a security method, and the absence of a symbol to having no security. If it shows native code and certified instead as symbols, with no symbol for managed, then beside being misleading re 1. and 3., it's not structured to show any particular category of information at all - it's a hodge podge. In fact, non-technical users will probably come away thinking that "native code" is a form of security. This is very, very bad design. Read Edward Tufte.

6. Again, to show how poorly "native vs certfied" works, iPhone will be switching to a certfied model in the production release of the SDK. (Probably; some form of security is promised, and that's the easiest to add.) Obviously the code won't have changed from ObjC to Java or Lua, so it will still be native, although it will be certified!

Umptious (talk) 20:14, 17 April 2008 (UTC)

I think your view of the average user's priorites are incorrect: People are generally interested in whether they can get third-party applications with unrestricted access to hardware. On what basis do you think that the average user is curious to know whether or not the Final security model is known?

-- End users don't even know what "third-party applications with unrestricted access to hardware" means. So they are not interested in knowing whether a platform has this by definition. Otoh they are interested in "Can my phone be used to steal my PayPal password?" "Can Fred's phone be used to create a cyberterrorist attack"? More: running in native code does NOT necessarly provide the unrestricted access to hw you imagine. For instance, if you check the spec of the Nokia N800 you'll find it has abuilt-in PowerVR chip. This can't be reached, even though the platform allows unrestricted native code. Otoh, a manufacturer using the same chip and permitting only Java to be run could provide access if he chose. The issue that you imagine is important to you is one that is much more complex than you think. Umptious (talk) 14:53, 18 April 2008 (UTC)

Please leave in the native code specification. InternetMeme (talk) 04:23, 18 April 2008 (UTC)

On the other hand, your addition of the "capability-based security" specification is very good ; ) So thanks for adding that! InternetMeme (talk) 04:28, 18 April 2008 (UTC)

I've re-read the information you provided, and realized that we've probably got two different equally valid perspectives on operating systems:

  • You're thinking along the lines of "How secure is this platform? What security policies does it enforce? Does it restrict access to the hardware? Or will it allow native code to run uninhibited?"
  • I'm thinking "How similar to a regular desktop operating system is this? Can I get full access to the hardware? Can developers write procedures in assembler to enhance performance? Will it perform?"

- Your POV only to a hobbyist whose aim is to use the platform to write C code - because that *is* his hobby. A consumer, analyst, manager, political decision maker, or citizen concerned at data theft or cyberterrorism will want to know about security. The first is, I've said, a specialist interest for a hobby web site, the second is appropriate for a general encyclopedia. (A *professional* programmer's interests will be different again - but security will probably be the most important: switching being native and managed code comes easily to a pro, but if a platform is fundamentally insecure he may have to abandon it.) Btw, if you think writing assembly automatically corresponds to increased performance today then you should read a good book on modern processor architecture and optimization. (Try Dowd's book, or Randall Hyde's, especially if you still think that the cpu cycle per instruction time in the handbook really determines performance...) And I say this as probably the only person who managed to get 16bit realtime gouraud shading running on a non-mmx Pentium in a real game - which let me double my salary and got me invited to Intel game dev conferences and one of the first half dozen MMX PCs in Europe, back in the day). Umptious (talk) 14:53, 18 April 2008 (UTC)

Basically, the "native code" tag is completely unrelated to security. It's just there to let people know which operating systems allow more unfettered access to hardware.

--InternetMeme (talk) 13:31, 18 April 2008 (UTC)

- Having "native" as am extra category may not be harmful, as long as it used correctly. I'll think about it. But you've marked several platforms incorrectly. Mobile does have a cert system - although I think it can probably be ignored; iPhone won't be allowing open native code access in the final developer SDK (see the article I linked from the Maemo page); Qtopia only allows approved middleware in run in native - general third party apps are managed, so it's only native in the minimal sense that *any* os has to be: check the platform docs linked from the wiki article. I'll come back and correct these when I have time. Oh, and leave a message on my wiki page if you want more concrete book suggestions. Umptious (talk) 14:53, 18 April 2008 (UTC)

Just added native code indicator for LiMo; I'm also editing the LiMo Platform article, with a reference there. 129.74.64.45 (talk) 00:18, 25 April 2008 (UTC)

[edit] navbox vs infobox

Would this template fit better as a {{navbox}} instead of an infobox? Here is a quick mockup:

v • d • e
Mobile Operating Systems
l = Linux based · m = Managed code support · n = Native code permitted · s = Capability-based security

Many of the linked articles already have an infobox and this gets pushed below it, which messes up some of the section [edit] links directly below the table of contents. See iPhone OS to see what I mean. ~ PaulT+/C 23:51, 20 April 2008 (UTC)

Hmmm. Well, I think it's easier to read with all the items in a vertical list. Articles such as DVB and GSM have a list of standards in a similar format, at the vertical layout is very easy to read. navboxes are nice, but I think it's easier to read in its current format.
Nice job of the navbox, though : )
InternetMeme (talk) 16:11, 21 April 2008 (UTC)
I strongly agree with this change. Navboxen should look consistent. We should convert this one before it's transcluded too widely to make it easy to do in future. Chris Cunningham (not at work) - talk 16:28, 26 April 2008 (UTC)
I've seen another list like this called Template: Table Mobile phone standards that works rather well. It's easy to access and browse through a few items on the list without scrolling down to the bottom of the page each time; it's easier to read, as a vertical layout works well with lists. Also, in the vertical format, it can easily be transcluded in context within sub-sections of other articles pertaining to operating sytems. We could always have a navbox at the bottom as well : ) InternetMeme (talk) 17:04, 26 April 2008 (UTC)
The Mobile Standards box works because unlike operating systems, networking standards typically don't have navboxes at the top of their articles, nor particularly many screenshots. In this case the template clashes with other templates on its transclusions; it doesn't play well with {{ambox}}es on at least Firefox 3, and stacks poorly with infoboxes. As for having both, that'd just lead to even more link-spamming. Chris Cunningham (not at work) - talk 17:13, 26 April 2008 (UTC)
Hey, I'd better sleep on it (It's very late here). I'll write the reasons I think the Table is better when I get up : ) I'll be back soon. —Preceding unsigned comment added by InternetMeme (talk • contribs) 17:20, 26 April 2008 (UTC)
Hey, well, Independently of what's a better idea out of the infobox vs. the navbox, it's definitely a good idea to keep them separate, as they have different names (Template:Mobile operating systems vs Template:Table mobile operating systems).
Both formats have pros and cons. For instance, the current format is suitable for transcluding inline in certain types of articles. I agree that navboxes are also goo though, so why not make one of them as well? If the people that are involved in the respective articles prefer one over the other, then they can take their pick : ) Or use both, if it's a large article : ) InternetMeme (talk) 12:46, 3 May 2008 (UTC)
Because that's a poor compromise. I don't feel that there's any reason to not use a navbox in all cases, so the end result would be the same (replacing all the templates). You haven't shown which "certain types of articles" the floated template is more suitable for. Chris Cunningham (not at work) - talk 13:03, 3 May 2008 (UTC)
Well, from my point of view, I don't see why we shouldn't just leave it the way it is in all cases, so the compromise seems sensible to me. It seems as though the only advantage to changing it is that it will match a system that may not feature the best layout. Wikipedia is constantly evolving in all respects—layout included—and trying new ways of presenting information can be a good thing. Also, Template:Infobox_OS doesn't play well with {{ambox}} either, but it would be a bad move to delete that.—Preceding unsigned comment added by InternetMeme (talk • contribs)

This doesn't work well as an infobox because most of the pages it links to and is transcluded on already have infoboxes themselves. This results in a messy situation with two infoboxes one right after the other. As I mentioned in the first post of this section, this can cause formatting problems with section [edit] links. Also, many of these pages are also very short and having a long infobox like this looks out of place.

Converting this list to a navbox would mitigate many of these problems. I haven't seen any concrete arguments for why it should be kept as a list other than "it looks better to me that way" and "there are other infobox templates similar to this one". Neither argument addresses the issues that have been pointed out in this discussion. ~ PaulT+/C 19:17, 14 May 2008 (UTC)

[edit] Inclusion of maemo

Should maemo be included in this list? It seems to me it should be, as the term "mobile" doesn't seem to be used as shorthand for "mobile phone" here, but there's nothing clearly stating that non-telephone devices are included, and all other OSes listed are (or are planned to be) used in phones. —Preceding unsigned comment added by 129.74.64.45 (talk) 00:36, 25 April 2008 (UTC)

I agree : ) Check the "Internet Tablet OS" link. InternetMeme (talk) 03:28, 25 April 2008 (UTC)

I disagree in part, if we are going to include ITOS, why not eeePC, Windows XP, and Windows Vista as they are used on other physically "mobile" devices. Unfortunately there doesn't appear to be a clear naming consistency in the industry to rely on in this case. Brontide (talk) 13:17, 25 April 2008 (UTC)

Yeah, Brontide, I think you're on to something. And as time goes on, desktop operating will be used on more mobile devices, and conversely, an increasing number of mobile operating systems will rival desktop operating systems in functionality; and the line will blur still further. Probably this issue will have to be addressed in future, but for now, my general definition of a mobile operating system is something like "Any commonly used or well-known operating system designed to work on a device that fits in an average sized pocket". Not a great definition, but its a start : )

InternetMeme (talk) 13:29, 25 April 2008 (UTC)

[edit] LiMo Platform

Is the LiMo Platform itself an OS? I see someone added MotoMagx below it. I assume that is an OS based on the LiMo Platform? Why have both listed? Shouldn't just the OSes be listed with a note saying which ones are based on the LiMo Platform if that is relevant? ~ PaulT+/C 19:04, 14 May 2008 (UTC)