Talk:Windows Server 2012/GA3

GA Review

Article (edit | visual edit | history) · Article talk (edit | history) · Watch

Reviewer: Retrolord (talk · contribs) 02:56, 30 January 2013 (UTC)

Article put on hold due to:

Some NTFS features are not supported in ReFS, including named streams, object IDs, short names, file compression, file level encryption (EFS), user data transactions, sparse files, hard links, extended attributes, and disk quotas.

Appears to be copied from source 34. I failed the article last time for systemic plaigirism, please check the sources in this article, I have only checked two references (one was in teh last review) and both times i came across plaigirism.

This is not a direct copy because "including" in the article is not the same as "specifically" in the source. I've checked the other sources and I'm sure there's no plagiarism, if there is, I will remove it.--Jasper Deng (talk) 03:07, 30 January 2013 (UTC)

Im not sure that this warrants mention in the article: The Australian construction company Kennards was impressed by the stability of the OS.[53] — Preceding unsigned comment added by Retrolord (talk • contribs) 03:11, 30 January 2013 (UTC)

Despite your claims i still find that the following section is plaigirism.

Note the similarities:

named streams, object IDs, short names, file compression, file level encryption (EFS), user data transactions, sparse files, hard links, extended attributes, and disk quotas.[35][34] (from the wiki article)

named streams, object IDs, short names, compression, file level encryption (EFS), user data transactions, sparse, hard-links, extended attributes, and quotas (from http://www.zdnet.com/blog/microsoft/microsoft-goes-public-with-plans-for-its-new-windows-8-file-system/11666)

Please rectify this.Retrolord (talk) 03:16, 30 January 2013 (UTC)

Further plaigirism

This sentence:

symbolic links, junction points, mount points, reparse points, volume snapshots, file IDs, and oplock. ReFS seamlessly[35]

Is copied from the source, also, the source is in the wrong spot. — Preceding unsigned comment added by Retrolord (talk • contribs) 03:19, 30 January 2013 (UTC)

Further plaigirism

This sentence:

IPAM provides for administration and monitoring of servers running Dynamic Host Configuration Protocol (DHCP) and Domain Name System (DNS).

Copied from http://technet.microsoft.com/en-us/library/hh831353.aspx Retrolord (talk) 03:28, 30 January 2013 (UTC)

Excessive referencing

This sentence:

Up to 1024 virtual machines can be active per host, and up to 8000 can be active per failover cluster.[29][31]

I am unsure as to what [29] has to do with it. Retrolord (talk) 03:30, 30 January 2013 (UTC)

Look, single listings like these are not copyrightable to this extent. I've removed the redundant reference. As long as it is sufficiently paraphrased it is not plagiarism.--Jasper Deng (talk) 03:37, 30 January 2013 (UTC)

Yes, if they were paraphrased they would not be plaigirism. But these are not paraphrased.

Retrolord (talk) 03:38, 30 January 2013 (UTC)

(please use indenting for readability) No, the sentences as a whole are paraphrased. Besides, anyone can say these items in that order.... does that make it a copyright violation? No. If you have big problems with that, change the order of the list accordingly, but I feel that this is not considered plagiarism.--Jasper Deng (talk) 03:43, 30 January 2013 (UTC)

Problems with the intro:

• The software was generally available to customers starting on September 4, 2012 and worldwide through multiple channels in September 2012.[3]

What does that mean?

• Although it has a Metro user interface (which has led to a mixed reception for Windows 8), the operating system has generally received positive reviews for these features.

Implies Metro user interface is a bad thing?

• Various features were added or improved over Windows Server 2008 R2, such as an updated version of Hyper-V, an IP address management role, a new version of Windows Task Manager, and ReFS, a new file system.

Implies it is better than Windows Server 2008 R2, is this the broad consensus? It mentions additions and improvements, but the metro user interface is mentioned in a negative way, so shouldnt you clarify that not everything was good?

Please work on these points. Thanks, Retrolord (talk) 03:45, 30 January 2013 (UTC)

• • I've been asked to take a look at the plagiarism concerns. "IPAM is used for administration and monitoring of servers running Domain Name System (DNS) and Dynamic Host Configuration Protocol (DHCP)." is a bit too close to the source for my tastes, but Jasper Deng is right in that you can't copyright a list like that since the items can't be paraphrased as they are what the technology is called. --Rschen7754 03:48, 30 January 2013 (UTC)

Thanks for the help Rschen7754! I'll give up on the lists, but as you said the IPAM part is too close to the source. Retrolord (talk) 03:50, 30 January 2013 (UTC)

• The first sentence is only intended to summarize. The history section and the editions section clarify it. The second sentence addresses the notion that because the OS is based off Windows 8, people might criticize it for the Metro UI as well - the sentence says nothing about the interface. The third one is clearly shown by the reception and the list of changes.
• Mentioning the construction company says something about the stability of the OS, something readers are looking for.--Jasper Deng (talk) 03:53, 30 January 2013 (UTC)
• I've fixed the IPAM sentence. As for the Metro sentence, you bring up a good point. Wikipedia is about providing due weight proportional to coverage in reliable sources. "Generally positive" does not mean "exclusively positive", and a few sentences on the reception sentence mention such criticism.--Jasper Deng (talk) 03:56, 30 January 2013 (UTC)

The sentence: The software was generally available to customers starting on September 4, 2012 and worldwide through multiple channels in September 2012.[3]

Does not make sense, it was generall available on sept 4, and worldwide in september? If it was already generally available on sept 4, why was it then released worldwide. i dont understand.

Also, is a construction company qualified to comment on the stability of a server OS? I would prefer it if you got another source for that point, from somewhere qualified in such matters. Retrolord (talk) 03:57, 30 January 2013 (UTC)

Since it is mentioned the metro interface in the lead and implied it is bad, could you add some bits into the reception section regarding this? and is there a broader consensus that metro is bad? Retrolord (talk) 04:00, 30 January 2013 (UTC)

(edit conflict)A large construction company has many computers, and as the source points out, its virtualization workload tests the ability of this OS. It is very unfortunate that I could not find any other third-party reliable sources about such case studies - all the other case studies I found while Googling it were from Microsoft's own website. I got rid of the confusion in that sentence because I now feel it's not worth clarifying the actual release schedule.

It does not imply Metro is bad. It only implies that it was a point of criticism.--Jasper Deng (talk) 04:08, 30 January 2013 (UTC)

• Regarding the construction company line, i must insist it is removed as unless you can prove they are qualified to make such statements, we cant assume its a reliable source. You dont know what the workload of that companys computers is , and anecdotal evidence is not a reliable source.

• Please get sources proving metro was a source of critiscm.

• Do we need the phrase "(which has led to a mixed reception for Windows 8)", perahps a note at the end of the article would serve this purpose better.

Retrolord (talk) 04:12, 30 January 2013 (UTC)

I feel this sentence could be worded better.

Microsoft introduced Windows Server 2012 in BUILD 2011 conference on 9 September 2011; Windows Server 2012 developer preview was released on the same day.

Retrolord (talk) 04:16, 30 January 2013 (UTC)

• ReFS resiliency features enhance the mirroring feature provided by Storage Spaces and can detect whether any mirrored copies of files become corrupt using background data scrubbing process, which periodically reads all mirror copies and verifies their checksums then replaces bad copies with good ones.

and

• ReFS resiliency features enhance the mirroring feature provided by Storage Spaces and can detect whether any mirrored copies of files become corrupt using background data scrubbing process, which periodically reads all mirror copies and verifies their checksums then replaces bad copies with good ones.

Sources?

Retrolord (talk) 04:17, 30 January 2013 (UTC)

• If they run virtualization workloads on a medium-to-large scale, they're qualified. Read the source.
• Please read the Reception section more carefully. It does mention criticism of Metro.
• I think we do, because otherwise the reader doesn't know the purpose of that sentence.
• Fixed to be more concise.
• The citations are at the beginning of the section. If you feel that this is a big issue please feel free to add additional citations to those sources as necessary.--Jasper Deng (talk) 04:21, 30 January 2013 (UTC)

• • It is anecdotal evidence, you dont know who in the company made the comment or what qualifications they have to judge it as "stable". An OS as big as this cannot be judged stable or unstable based on one piece of anecdotal evidence. This article will NOT be passed until i am satisfied on this issue, so please back up this claim
  • I am unable to find any critiscm of metro specfically
  • Sentence does not read properly, i think it should change but happy to hear your opinion
  • Dont worry about the citations

Retrolord (talk) 04:28, 30 January 2013 (UTC)

• • • Please read the source again. The first paragraph mentions who said it. The judgement is by an IT department, a firm they hired.
    • Because simply mentioning criticism is due weight based on the coverage it got.
    • Take another look, you may have missed my latest edit, or if you didn't, make suggestions, because I think it's pretty concise as written.--Jasper Deng (talk) 04:33, 30 January 2013 (UTC)

• Microsoft introduced Windows Server 2012 and its developer preview in BUILD 2011 conference on 9 September 2011.[7]
• Even if the comments were made by the IT department, it is still anecdotal, to back up such a claim you would need multiple references and be able to prove that only a minority believed it was unstable.

Retrolord (talk) 04:50, 30 January 2013 (UTC)

• • Sentence fixed.
  • I tweaked it to clarify that it's only their opinion.--Jasper Deng (talk) 04:52, 30 January 2013 (UTC)

• Windows Server 2012 improves over its predecessor Windows Server 2008 R2:

Sentence seems biased against predecessor? Not sure thats appropriate.

• Is scalabiltiy an appropriate heading for this section?

• The new version contains new features such as CPU usage caps for particular websites,[38] centralized management of SSL certificates, and improved support for NUMA, but little else has changed.[39]

Not sure about that sentence, especially the little else has changed part, not encyclopedic tone. Retrolord (talk) 04:58, 30 January 2013 (UTC)

Not sure this is notable enough for mention?

The new task manager recognizes when a Windows Store app has the "Suspended" status. Retrolord (talk) 05:05, 30 January 2013 (UTC)

• Scalability is the most concise title. It might be justified to move it into the system requirements section, but I don't see this as a big issue.
• The "little else has changed" is in accordance with the source, though I will probably change it to "no other substantial changes were made."--Jasper Deng (talk) 05:14, 30 January 2013 (UTC)
• It is an important feature of the new task manager. "improves over its predecessor" is justified because of the higher numbers, implicitly only in this aspect (and not in others).--Jasper Deng (talk) 05:18, 30 January 2013 (UTC)

Are you sure this section is not too technical for wikipedia?

ReFS uses B+ trees[34] for all on-disk structures including metadata and file data. The file size, total volume size, number of files in a directory and number of directories in a volume are limited by 64-bit numbers, which translates to maximum file size of 16 Exabytes, maximum volume size of 1 Yottabyte (with 64 KB clusters), which allows large scalability with no practical limits on file and directory size (hardware restrictions still apply). Metadata and file data are organized into tables similar to relational database. Free space is counted by a hierarchal allocator which includes three separate tables for large, medium, and small chunks. File names and file paths are each limited to a 32 KB Unicode text string. Retrolord (talk) 05:16, 30 January 2013 (UTC)

• Not too technical - anyone who is interested in the file system section should understand the units. Consider the target audience for this particular OS.--Jasper Deng (talk) 05:18, 30 January 2013 (UTC)
• It seems fine - granted, I have a degree in computer science, but a lot of our articles get fairly technical. --Rschen7754 05:20, 30 January 2013 (UTC)

• As of 23 September 2012, all students subscribed to DreamSpark program can download Windows Server 2012 Standard or Datacenter free of charge.[13]

It is still correct, and it's worth mention because it's one way the software is distributed. Remember, good articles must have broad coverage.--Jasper Deng (talk) 05:23, 30 January 2013 (UTC)

Does this warrant mention in the article? If you think it does, please also verify that it is still correct. Retrolord (talk) 05:20, 30 January 2013 (UTC)

• The new task manager recognizes when a Windows Store app has the "Suspended" status.

Despite what you said, i dont think this is an important feature of task manager. i dont think it belongs on a page about windows server 2012. Retrolord (talk) 05:22, 30 January 2013 (UTC)

No it is. It was one of the most significant changes in Task Manager - the ability to simply pause running programs.--Jasper Deng (talk) 05:23, 30 January 2013 (UTC)

If that is the case then the sentence doesnt convey that meainng. To me it meant task manager now tells you if windows store is "suspeneded". Didnt come across as, you can now pause programs whenever you want.

Retrolord (talk) 05:26, 30 January 2013 (UTC)

It does convey the meaning in a concise manner, as "Windows Store" is a descriptor for "app". The "Windows Store" designation is out of lack of a better choice because Metro is only the user interface.--Jasper Deng (talk) 05:30, 30 January 2013 (UTC)

• Unlike its predecessor, Windows Server 2012 has no support for Itanium-based computers

Please explain why this is in the lead. I see no mention of it anywhere else in the article so i am curious. Retrolord (talk) 07:23, 30 January 2013 (UTC)

It is mentioned, albeit indirectly, in the system requirements section ("only runs on x64 processors"). It was Microsoft's decision, and I don't feel like it's worth elaborating on why they chose - they choose as they do.--Jasper Deng (talk) 07:41, 30 January 2013 (UTC)

If it is not deemed suitable to be referenced directly at all in the article i would argue it isnt worth including it in the lead then. Retrolord (talk) 07:45, 30 January 2013 (UTC)

I included it, since it is a significant shift for Windows Server. Also, before you make your final verdict, please ensure you are following the good article criteria point-by-point.--Jasper Deng (talk) 07:47, 30 January 2013 (UTC)

Back to the issue of the australian construction company. I have decided that it is both anecdotal and an unreliable source. The article you cited as "proof" that windows server 2012 is stable is written like an advertisement. Furthermore, I am inclined to believe that ZDNet is NOT a reliable source for information relating to Microsoft. This is based on http://www.roughlydrafted.com/2007/08/24/zdnets-george-ou-exposed-as-ignorant-microsoft-shill-zoon/ and as a result I cannot accept that site as a reliable source.

To address your request for me to follow the the good article criteria point-by-point, I intend to conduct a thorough review of this article and will follow those criteria. Retrolord (talk) 08:32, 30 January 2013 (UTC)

No, ZDNet's authors post highly reputable information. Notice that the article does not claim that the OS is stable for everyone. It only claims that the company found it to be stable for their purposes, which is definitely backed up by that source. And anyways, that link is to a WP:BLOG that only contains the personal opinion of a critic of that author. On the other hand, the ZDNet article makes no outstanding claims. It is written professionally (unlike the blog you linked) and I have no reason to not trust it. ZDNet has been cited countless times in Microsoft-related articles, so it is definitely reliable.--Jasper Deng (talk) 08:39, 30 January 2013 (UTC)

In the IT managers opinion, The OS is stable. This is both an opinion and aneccdotal. The burden of proof lies with you to prove that the OS is stable if you are to include such claims in the article. Regardless of how reliable ZDNet has been on other occasions, just because something is posted there does not mean it is fact.

I also request that you write a section detailing the Microsoft's changes since the last version of the OS and directly reference the lack of itanium processor support. Retrolord (talk) 08:46, 30 January 2013 (UTC)

This is why it's not presented as a fact! It is a fact that the company IT manager has that opinion. That opinion is not a fact.

Please see my latest edits to the article, where Itanium support being dropped is mentioned in the system requirements section. I don't think it's necessary to elaborate more on the new features because that's what this article essentially is mainly about.--Jasper Deng (talk) 08:47, 30 January 2013 (UTC)

"Regardless of how reliable ZDNet has been on other occasions, just because something is posted there does not mean it is fact." - Wikipedia is an encyclopedia, meaning that it is a tertiary source. We report what the sources say - we don't go do the research for ourselves. --Rschen7754 08:48, 30 January 2013 (UTC)

Did the company recieve this expensive OS for free? If so can they really be trusted to provide an unbiased report on the OS?

I just dont think its worthy of including in the article, basically, the article lists some notable sources with reviews, but then says effectively "this random company approves of this OS" Retrolord (talk) 08:51, 30 January 2013 (UTC)

They did not receive it for free (again please read the source carefully before you ask questions about it). Please note that reception sections contain the opinions of reviewers. It isn't just some random company - it's a company that has a large IT system, which is the kind of company that this OS is for. Whether they are "qualified" or not is irrelevant. If I could, I would include other company's case studies but as I said before this was the only third-party reliable source I could find.--Jasper Deng (talk) 16:41, 30 January 2013 (UTC)

Please explain the sources for all information in the color chart. I have to ask this as I have TWICE uncovered plaigirism in this article so feel it neccessary to thoroughly inspect the article. Retrolord (talk) 22:20, 30 January 2013 (UTC)

This comparison chart is from information from multiple sources. Like a list of items this table's information is not copyrightable. --Jasper Deng (talk) 00:49, 31 January 2013 (UTC)

I did not mean to imply that the charts had been plaigirised, i was just noting the two previous incidents of plaigirism which lead me to believe the article was not written as thoroughly as it could have been, while still assuming good faith in the actions of all contributors.

I am curios as to the source of several statements in the chart, which after nominally reviewing the references provided could not see anything relating to "Fax server Role" or "Print and document services". Could you please point out where these two assertions come from,

Thanks! Retrolord (talk) 07:11, 31 January 2013 (UTC)

Please see source #45.--Jasper Deng (talk) 07:26, 31 January 2013 (UTC)

Also, before you jump to conclusions, please explain exactly how you feel the article is not written thoroughly.--Jasper Deng (talk) 07:27, 31 January 2013 (UTC)

The two incidents of plaigirism lead me to believe that.

I noticed in the source 45 you mentioned that Datacentre and standard seem to be grouped together on the charts in that PDF. Please explain why you have taken a different approach in this article?

Should they be combined as they are in the references you are using for consistancy? Retrolord (talk) 08:11, 31 January 2013 (UTC)

You cannot jump to that conclusion without actually examining the content and layout of the article.

No, they should not be combined, as that would complicate the table. They have one feature difference but it's major - virtualization rights. They also differ by distribution methods available.--Jasper Deng (talk) 08:14, 31 January 2013 (UTC)

Windows 8 and Windows Server 2012 include a new version of Windows Task Manager together with the old version.

Reference to Windows 8 should be removed? Retrolord (talk) 08:23, 31 January 2013 (UTC)

Done although I do not believe this is a big issue at all. Tip: instead of pointing out the issues one-by-one, please list them all at once. This is more efficient.--Jasper Deng (talk) 08:26, 31 January 2013 (UTC)

On that last point, i dont see how source 17 relates to that sentence or where in the source it backs up that claim.

Also, this sentence: In the new Processes tab, the processes are displayed in varying shades of yellow, with darker shades representing heavier resource use. After reviewing source 18, which i assume refers to this whole section, i cant see anything in the source to back up that claim? Retrolord (talk) 08:30, 31 January 2013 (UTC)

Source 17 is needed because we need to back up the claim that the old one still exists. Source 1920 clearly shows yellow heat maps (although I may want to get another source). Another tip: use colons to indent for better readability.--Jasper Deng (talk) 08:36, 31 January 2013 (UTC)

(edit conflict)Added another source to clarify.--Jasper Deng (talk) 08:41, 31 January 2013 (UTC)

Please reformat the section so it is clear what each reference is referring to, having the claim about yellow things at the top and source [19] for them at the bottom doesnt make sense and makes it difficult for the reader to validate the claims. Retrolord (talk) 08:40, 31 January 2013 (UTC)

(done, please use this indenting because it's hard for others to follow our discussion otherwise).--Jasper Deng (talk) 08:41, 31 January 2013 (UTC)

Well im satisfied with the article then. Sorry i've had to be so thorough, I admit i've been much more stringent than usual. It was unfortunate that an editor other than yourself had taken it upon themselves and to plaigirise material, a regrettable act.

Thankyou for cooperating throughout the entire process, you have been quite helpful throughout my review. Thankyou, your article has passed. Retrolord (talk) 08:54, 31 January 2013 (UTC)