Talk:Comparison of real-time operating systems

	This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing articles
Low	This article has been rated as Low-importance on the project's importance scale.
	This article is supported by WikiProject Software (assessed as Low-importance).

Licensing errors

RTX Keil is mentioned as open source however only the CMSIS headers really are. This is the same as saying Windows is open source since Microsoft distributes headers. All of the RTX and CMSIS actual implementations are proprietary. Even RTX Keil's homepage asks for a quote request for their proprietary software: https://www.keil.com/arm/rl-arm/kernel.asp and even when you have a license to MDK, you don't get the license for the chip manufacturer's implementation. Kirliyarna (talk) 20:35, 20 September 2022 (UTC)[reply]

Attribute expansion ideas

Add performance. Is RTOS deterministic? Typical latency in CPU cycles. — Preceding unsigned comment added by Jharris99 (talk • contribs) 15:41, 9 July 2016 (UTC)[reply]

Agree, RTOS-specific characteristics could be added. For now the list is a basic comparison from the outside, which I would be fine with for an encyclopedia. Deep-comparison is always tricky. Sometimes you need to contact the vendors to find out what the OS is capable of because the normal data sheet won't tell you. (I have done such things.). I'd say the list (table) should stay like this but should be updated. There are a number of RTOS missing and I guess links need to be re-verified if working etc. KR 17387349L8764 (talk) 10:15, 20 September 2021 (UTC)[reply]

More columns for

royalties and fees
open source or closed source
approximate deployment numbers
initial date of development

—EncMstr (talk) 21:47, 21 May 2008 (UTC)[reply]

Good idea. What about these? (see also: Template:Infobox OS)

Company/Developer
Programmed in
Notes/Comments

Ghettoblaster (talk) 22:40, 21 May 2008 (UTC)[reply]

What about:

Latest release
Latest release date

This should help to sort out active projects from old/dead projects or systems interesting only as historical reference.

N'SallaNuto (talk) 13:36, 22 May 2008 (UTC)[reply]

The release number often creates a quagmire of updates. Anyway, the status column is intended to reflect whether it is still supported and under development. —EncMstr (talk) 17:58, 22 May 2008 (UTC)[reply]

TOPS-10

I do not see TOPS-10 listed. It was used in some very time critical applications. One project was control of the north west power distribution. Bringing power stations on line and off line sinking them up to the grid. In 1972 DEC listed "real time processing" as one of TOPS-10 features. It provided locking of tasks in memory and provided 10 high priority run queues. If there is a reason it's not listed pleas explain. Steamerandy (talk) 12:17, 11 January 2013 (UTC)[reply]

Hi @Steamerandy, fine with me, add it? Many more RTOS not in this list and links probably need updating throughout the document. KR 17387349L8764 (talk) 10:11, 20 September 2021 (UTC)[reply]

external links from original article

This hive of external links used to be in the original article before the split; they should be incorporated into the appropriate entries in this article.

—EncMstr (talk) 23:41, 22 May 2008 (UTC)[reply]

Are WindowsCE and SymbianOS really RTOS ?

I am not sure those can be classified as "RTOS", possible targets are embedded devices but this alone does not give them real time capabilities. Probably we should put them among the general purpose operating systems.N'SallaNuto (talk) 08:06, 24 May 2008 (UTC)[reply]

Yes, operating systems that target embedded devices are often incorrectly described as "RTOS".

However:

The SymbianOS article says "Later OS iterations ... notably the introduction of a real-time kernel" and "Symbian OS EKA2 supports ... real-time response".

The Windows CE article says "Windows CE conforms to the definition of a real-time operating system".

I think an OS whose latest version is real-time should be listed here as a real-time operating system, even though the earliest versions were not real-time.

If you have some evidence that these OSes are *not* real-time, please put it in those articles and specifically state " ... is not a real-time OS ...". But as long as the article for an OS claims it is "real-time", it needs to stay on this list.

--68.0.124.33 (talk) 14:39, 19 June 2008 (UTC)[reply]

I agree with anon IP. Ghettoblaster (talk) 23:56, 23 June 2008 (UTC)[reply]

I searched the web for some info regarding WinCE RT capabilities, I found some documentation and it seems that since version 3 it supports priority inversion avoidance mechanisms and other features required for a RTOS. Probably a note about this (version 3 or above) should be added to the article and/or the list. N'SallaNuto (talk) 08:13, 24 June 2008 (UTC)[reply]

OSEK

OSEK is a standard for a certain type of operating systems, not an actual implementation... (compare with POSIX for traditional operating systems) TERdON (talk) 15:50, 15 October 2008 (UTC)[reply]

There is another entry in the table like that too, though I don't immediately see it. However, I've taken the liberty of filling in the table appropriately. Comments? —EncMstr (talk) 16:25, 15 October 2008 (UTC)[reply]

Yes, OSEK is a standard (I prefer calling it specification), and many variants of the OS exist. Updating article, closing this comment section.

Done 17387349L8764 (talk) 07:19, 21 June 2021 (UTC)[reply]

ARTOS: duplicate names, different operating systems

Some operating systems are entirely different but use the same name. For example "ARTOS" is used by Locamation, and by Mike Fowler for their respective OSes, (and I believe a third company as well). But these OSes are as different as Windows CE and Symbian OS.

At the moment these two differnt OSes are merged to the same line. In my opinion these should be on separate lines. Andrec79 (talk) 08:03, 15 December 2009 (UTC)[reply]

Absolutely. If you know enough about them, please generate unique names for them and divide the entry into multiple lines accordingly. Something like ARTOS (mfg name) and ARTOS (whatever). —EncMstr (talk) 18:04, 15 December 2009 (UTC)[reply]

Linux RT patch

Where is the "Linux RT patch" (PREEMPT_RT) by Ingo Molnar? —Preceding unsigned comment added by 84.72.31.5 (talk) 08:28, 22 November 2010 (UTC)[reply]

It's in the list now, search PREE*

Done 17387349L8764 (talk) 07:20, 21 June 2021 (UTC)[reply]

As of Linux 6.12, PREEMPT_RT has been merged and Linux has officialy real time capabilities. — Preceding unsigned comment added by 81.187.56.149 (talk) 23:01, 18 November 2024 (UTC)[reply]

Non-notable entries

There seems to be a slow edit war involving removing/reinstating RTOS entries in bulk, apparent on notability grounds. While the notability policy applies to articles, it does not apply to individual entries in an article. For example, many athletic sport articles which do not list yearly winners would be considered incomplete even if each winner and competition was not notable. —EncMstr (talk) 18:50, 20 August 2012 (UTC)[reply]

Unfortunatly as it relates to non article entries, this is guideline applies to lists, The Purpose of Lists in wikipedia is used for internal navigation and "...lists which exist primarily for development or maintenance purposes (such as a list that consists primarily of red links) should be in project or user space, not the main space. ". Wikipedia has 6,917,947 and is optimized for readers over editors, redlinks and non article entries are unhelpful to readers. Non-article entries do not add content or meaning to the encyclopedia. Please Write the Article First. Additionaly, "Lists" are subject to Wikipedia's other policies such as WP:NOTDIR and WP:SPAM. Equally, Wikipedia is not a guide nor a repository to any/every real-time operating system in existence. --Hu12 (talk) 17:35, 2 November 2012 (UTC)[reply]

Hu12, your repeated bulk deletions and the attempt to completely redefine the established purpose of this article over the past months without seeking any consensus despite being reverted many times by other editors made it increasingly difficult to still assume good faith. Therefore I hope, we can find a working consensus now.

This article was and is a list of real-time operating systems, not a list of particularly notable (by WP's definition) real-time operating systems. Since there never was nor is a requirement for individual entries in an article to be notable by itself, and a list is nothing but an article, it is okay to list operating systems, for which we don't have articles. I too would like to see more entries with articles, but Rome was not built in a day, so give it time. And even if some entries will remain without article forever, this is still okay by our standards, it still serves the purpose of the article.

While this article may also serve as a navigational aid, its main purpose is to give an informational overview and make quick comparisons possible, hence its table format. There is no guideline which would define that all articles containing lists must be for navigational purposes only (to the contrary) or contain only blue-links to existing articles. You are reading something into our guidelines and policies, which simply isn't stated there.

I am not going to defend every single entry in the list (I would not even mind to delete a few of them and add others, but other editors might not agree), but you even deleted entries, for which we do have articles, or which are otherwise well-known and in widespread use in the industry, thereby contradicting your own ruleset above.

While notability never was a criterium for individual entries in an article, stated facts should still be either obvious or verifiable, ideally by providing a reliable reference of some kind. So, I would agree to put a "more references needed" tag on this article in order to improve the quality of the various entries over time, but not to mass delete them.

There are some ten-thousand RTOSes in the wild. Therefore, the list in its current form and length does in no way risk to become a listing of every possible RTOS under the sun. There are still many more RTOS which would belong into here...

--Matthiaspaul (talk) 20:15, 2 November 2012 (UTC)[reply]

Improvements are needed to meet "The Purpose of Lists" Guideline, which is community agreed upon consensus and is considered a standard that all users should follow. I fail to see how it would be difficult to assume good faith when following the actual established purpose of Lists. Any editor would find attempts to bring this list up to community standards reasonable. The established purpose, is demonstrated in the following;

Per the Guideline WP:LISTPURP; "Navigation ...serve as natural tables of contents and indexes of Wikipedia... to assist in navigating their subjects,"
Per the Guideline WP:LISTPURP; "Development ...lists which exist primarily for development or maintenance purposes (such as a list that consists primarily of red links) should be in project or user space, not the main space."
Per the Guideline WP:RED; "...editors are encouraged to write the article first, and instead use WikiProjects or user spaces to keep track of unwritten articles."
Per the Policy WP:NOTREPOSITORY;"...merely being true, or even verifiable, does not automatically make something suitable for inclusion in the encyclopedia. "

This list, over the past four years has steadily degraded due to lack of maintenance, became a repository for any indiscriminate items, un-navigable and unhelpful for readers. Wikipedia is based upon collaborative, good faith editing, and consensus. I've attempted multiple times make the proper improvements, however it seems there is a case of WP:OWN by its creator imposing his own view of "standards to apply" rather than those of the community. The responsibility for justifying inclusion of any content rests firmly with the editor seeking to include it, I've seen no such reasonable or policy-compliant argument for these Non notable non-article entries to remain within this article. --Hu12 (talk) 04:22, 3 November 2012 (UTC)[reply]

You are overlooking the very first purpose in the guideline you mention, Purposes of lists, says: "Lists have three main purposes: 1. The list may be a valuable information source. This is particularly the case for a structured list. Examples would include lists organized chronologically, grouped by theme, or annotated lists."

This is the first and best purpose, and exactly describes what this article is. You have now asserted (twice) that the second purpose, "2. Navigation: Lists contain internally linked terms... serve(s) as natural tables of contents and indexes of Wikipedia..." and have altered the article to fit that description. And now you have added the third (and inferior) purpose from the guideline, an article development list.

This article serves multiple purposes and has worked well:

Comprehensively mention all significant RTOSs
Provide a summary and comparison of significant attributes of these products
Provide a means to find out more about each RTOS, and
Effectively deflect would-be spammers into providing structured, useful information as a resource to all who are interested in RTOSs. This article steadily averages 204 reads per day.

You may not be aware of the significant loads of spam (as WP:ELs) that the RTOS articles were receiving before I created this article four and a half years ago. Especially real-time operating system which was being hit at around ten times per week. At the time, it was an experiment for WP:WikiProject Spam, which I notice you are quite active in. (I was not active for long.) The experiment was wildly successful: it immediately decreased spam (of related articles) by at least 90%, (down to once or twice a month) and the edit rate here initially correlated to a deflection rate of 25–40%. That is, at least a quarter of would be spammers were improving this article.

Furthermore, the more apropos guideline is Wikipedia:Manual of Style/Stand-alone lists which states:

“

Common selection criteria: ...[second item] Every entry in the list fails the notability criteria. These lists are created explicitly because most or all of the listed items do not warrant independent articles: for example, List of minor characters in Dilbert or List of paracetamol brand names.

”

This is a community recognized guideline which states that every item in a list can be non-notable. Usefully collecting, organizing, and summarizing related information is encyclopedic per the Five Pillars. —EncMstr (talk) 07:56, 3 November 2012 (UTC)[reply]

Most certainly this warrants being a standalone list as there are enough articles that warrent inclusion. No question its been a great contribution that has prevented a lot of spam, thank you. However allowing content to be included simply because it meets the first criteria, while failing the other two is a slippery slope resulting in WP:NOTDIR. Precedent on "lists" that only meet the first of the three criteria is deletion¹². Any deletion discusion here would not end in deletion, however it would end up with this article being pruned and cleaned. You stated back in 2008 "I think that should address everyone's interests: keep the main article clutter free, provide a place for an exhaustive list which every RTOS developer wants, provide a check list for potential customers...". its not clutter free and seems to be incompatible with the aim of "The Purpose of Lists" Guideline. There is plenty of notable Articles in Category:Real-time operating systems all of which warrant inclusion here. Manual of Style/Stand-alone lists also states; "..lists devoted to a large number of redlinked (unwritten) articles—don't belong in the main namespace.". I fail to see why your against cleanup efforts, its not unreasonable and would make the article compliant like List of operating systems. --Hu12 (talk) 17:14, 3 November 2012 (UTC)[reply]

Nobody is claiming ownership of this article, Hu12. It is just that this article was developed with a particular (and valid) scope and it is you who wants to change that very purpose without consensus now.

As has been explained already, the article basically IS compliant with the relevant guidelines already (not in every little detail, but in general). There are certainly ways to improve the article, and everybody is welcome to put some effort into it, including you, of course. Improvements may also include some carefully crafted cleanup work.

However, what you did (and still propose) is not a cleanup, but a bulk deletion of lots of entries - entries, which serve the very purpose of this list and are perfectly okay to be listed here according to our community-approved guidelines, and, oddly enough, even some entries, which fulfill your own personal ruleset.

You attempted to completely reverse the established purpose and type of this list by intimidating other editors not to add new entries unless they meet your new personal criteria - the following has been added by you to the article:

"IF YOU DON'T PAY ATTENTION TO THIS MESSAGE, YOUR EDIT WILL BE ROLLED BACK WITHOUT WARNING. Only place entries here that are links to actual Wikipedia articles about notable real-time operating system. External links, redlinks, non-notable sites will be removed. If you have questions, use the talk page. Please try to keep entries in alphabetical order. Adding unnecessary links or text to any other section (such as the "References" section) will also be removed. Thanks."

Frankly, how do you come to make this bold change to the article without first seeking consensus with other editors? It's difficult to excuse it with a be-bold-revert-discuss editing style or someone just having a bad day, given that you didn't seek any discussion at all, even not after having been reverted many times over the course of months. Instead, you continued to show this editing pattern after I explictly asked you to seek a consensus here before deleting/modifying some contents again and you started to engage in an edit war with me. This would be inappropriate behaviour for any editor, but it is in particular for an admin.

I was wondering what drives you to exhibit this editing style and found these recent edits of Wikipedia:Write the article first, where you are bending the meaning of the essay to reflect your personal opinion rather than what was stated there before (in parts you almost reversed the meaning), again without an edit summary indicating your bold changes ("tweak") and without seeking consensus on the talk page first: [1]. From a cursory look at past discussions on this talk page it becomes clear, that your opinion is not even shared by many editors of that essay, and there is certainly no community consensus for it. I know, that the latter is not a requirement, but still essays should either not be biased or clearly indicate, that they represent a personal opinion only (which your edits fail to indicate), and they are "intended to supplement or clarify Wikipedia practices, guidelines, policies, or other Wikipedia norms and practices that in fact have communal consensus" and in fact that's how you use them to push something into a particular direction (see above).

Although I can still recognize and very much value your attempt to reduce SPAM (so much for the good faith), I find these actions highly questionable as they are going astray our goals. If being worn down by endless SPAM fighting causes such blind shots into the heart of the project as "collateral damage", it's perhaps not worth it...

Regarding your given precedent on lists that only meet the first of the three criteria, I don't see why lists would have to fulfill all criteria, providing information in a concise form is certainly the most important purpose of a list. TV channels are in multiple ways a completely different case, firstly, they are explicitly mentioned in the guidelines, and secondly, they are highly dynamic and will change every some days. A list of real-time operating systems is something very different, their relevance may change in the course of years or even decades.

Regarding red links, yes, there are some red links in the article, but by far not the majority of links is red, therefore I don't see them as a problem. There is no rule to not have red links in a list of this type. (BTW. Readers, who dislike red links for some reasons can configure the Wikipedia UI to not show red links as such.) But if red links are really such a problem for you, I propose to "unlink" some of the red links and just leave them as raw black text entries in the list (as it is supported by guidelines as well).

--Matthiaspaul (talk) 14:01, 4 November 2012 (UTC)[reply]

Matthiaspaul, I have adequetly explained my position and supported my position based on community supported policy rationale. EncMstr, has done the same. You yourself agree that improvements are needed here, however I fail to see how attempting to win arguments through attacking my motives with mischaracterization reaches that goal. With that said, I think we all agree that Guidelines such as WP:LISTPURP, WP:NOTREPOSITORY, WP:ED and WP:CSC are accepted as consensus among editors and are considered a standard that users should follow. I do agree to at least converting the redlinks to "text' base entries in this case. I was considering that myself, and am pleased you brought it up. thanks.--Hu12 (talk) 17:47, 4 November 2012 (UTC)[reply]

can some of the table headers/contents get an explanation? mainly "Status"

Hi there, it's not clear to me what the implications of some of the status types are. What is the difference between "closed", "defunct", "archived", "inactive'", "historic", and "discontinued"? If possible can someone add an explanation to the article? since I don't know why each status category was picked and what it means. — Preceding unsigned comment added by 68.36.216.10 (talk) 18:00, 24 May 2021 (UTC)[reply]

Hi, can you sign your comment please? The meaning could be documented, but what specifically are you unsure about? I would say the attribute is always difficult to fixate, it is hence a rough indicator. The most concerning part is, that it needs to be reviewed (the list) and up to date to have real meaning. Defunct is dissolved. Such things exist. Historic is old software, still functional, but maybe for old microcontrollers not in wide use any more or hardly available - such things exist very much. ... The other thing is, the attributes are to be combined. There might be inactive and close projects or software, meaning it has been inactive for a long time (no new releases, no fixes, no action at all) and they are closed by the team, but closed inactive only means it has not been touched for a long time, but also not formally closed, i. e. bad communication. KR 17387349L8764 (talk) 09:05, 21 June 2021 (UTC)[reply]