Talk:Time-Division Long-Term Evolution

Requesting help to improve this article

Hello, I've recently been working on behalf of Qualcomm to redraft this article. The current version suffers from a number of issues, including especially a lack of detail in some areas (for example, History) but excessive detail in others, along with a couple of bulleted lists where prose would be more appropriate.

Because of my financial conflict of interest, I won't make any edits to this article myself. Instead, I've uploaded my new draft to my userspace. By way of summary, the draft consists of the following sections:

• An introductory paragraph, which is loosely based on the current introduction, but edited to summarize the rest of the content that appears in my new draft.
• An "Overview of technology" section, which describes, in lay terms, what TD-LTE is and how it works.
• A "History" section, which discusses the development of TD-LTE but a variety of companies.
• Finally, a "Deployment, commercial networks and devices" describes the various networks and devices around the world which support TD-LTE.

If an editor has time to take a look at my draft and, if it looks okay, replace this article with that draft, I'd appreciate it! If you have any questions or comments, though, I'm all ears! Cheers, ChrisPond (Talk · COI) 15:17, 19 March 2014 (UTC)

I don't really know enough about the technical side of this to approve your draft. Your draft seems to de-emphasize TD-SCDMA, which, as I understand it, is mostly used in China but is important since it seems to have delayed wider LTE adoption, and is also time division like this LTE. Also, speeds are not mentioned here or in your draft, which seems to me to be one of the most important details to include. The draft seems a little cleaner, but without being able to detect subtle technological bias correctly, I'm reluctant to drop it in place. Gigs (talk) 18:13, 24 March 2014 (UTC)

Hey Gigs, fair enough—I can reach out to other editors to review the material. It's definitely a complicated topic, and one that took me a bit to get my head around while I was researching and writing. Regarding your two specific points, though:

• For TD-SCDMA, my understanding is more that TD-LTE was developed to be another LTE option, with evolution from TD-SCDMA in mind, but not as a motivating factor, so I mentioned it, but didn't discuss it in depth.
• Regarding speed, I did include this where specific speeds were mentioned, but most of the sources I have only talk about speeds in relation to very specific tests or trials, not in general terms like "TD-LTE is faster/slower than LTE FDD," so I included what I had in the History section. I think there's just too much variability in networks and deployment to get these kinds of sources (although admittedly that assumption is a bit WP:OR on my part).

Cheers, ChrisPond (Talk · COI) 21:48, 24 March 2014 (UTC)

It looks like TD-LTE will be slower in practice than FDD, because it only uses one band instead of two, but the advantage is that each carrier/phone only needs one band. From a little research I did, it looks like the underlying modulation and speeds are the same, but since the band is shared for the uplink, you can't be downloading and uploading at the same time (at least in terms of the actual radio). I don't have a specific source that says this, though. Gigs (talk) 18:38, 26 March 2014 (UTC)

RE: speed issues. If you find a source that makes concrete claims about speeds, there is a significant chance it is a non-credible source (our XXXX network has the fastest speed ever!!!1!). Wireless performance is subject to huge amounts of variables that impact practical performance. Older protocols had a number of stated theoretical maximums, but newer protocols theoretical maximums are so disparate from real world performance that most engineers are very wary of stating norms outside of controlled experiments. At least, that's how I was with my boss. Jay Dubya (talk) 16:37, 27 March 2014 (UTC)

Hi Chris, First, thanks for disclosing potential conflict of interest. As a non-expert, looking at the current version of the article and your draft, I'd say that your draft is a definite improvement over the current article, both in terms of readability and article structure. Offhand, I don't see any major neutrality or promotional issues specific to Qualcomm, which is the main thing we look for in COI reviews. I appreciate the issues Gigs has brought up, but wonder if these are show stoppers to adopting the new draft. If Gigs agrees that it is neutral enough and that the issues they raised could be worked out on the talk page of the article, then I would support adoption of the new draft. If not, I'll bow out, as I'm not qualified to assess the subtle technical issues. Thanks, --Mark viking (talk) 19:15, 26 March 2014 (UTC)

Hey folks, thanks for taking a look and for the extra poking around. I'm not opposed to adding something about speeds if we can find a source that lays it out without us having to pull WP:OR, but it looks to me, Gigs, like you're having the same issue I had—there's discussion of speed in various places, but no single sources lays it out very clearly. I'm also not at all opposed to folks working on the article after it's been placed; I don't view my draft as the final version of the article in any way. If there's agreement that my draft is an improvement, how would you both feel about moving it over, with the idea that there might be other details added, especially related to speed, in the near future (assuming we can find an appropriate source, of course)? ChrisPond (Talk · COI) 20:47, 26 March 2014 (UTC)

Hi. Chris contacted me via my talk page as I am a member of Project:Telecom. I am a networking professional; I've used wireless components extensively for a long period of time, and I have no financial relationship with Qualcomm or any other wireless manufacturer or VAR. I've reviewed the original draft and Chris' draft. I agree completely with the two prior editors that Chris' article is light years ahead of the current draft. SCDMA is a deprecated protocol; its worth pointing out it was deprecated prior to LTE; HSPA is a better fit for that mantle. In whole though, a larger brief of LTE's relationship with earlier protocol standards is valid. I am happy to take responsibility to get that side of things started. In the meantime, IMO Chris has established consensus here, his changes have no COI and in fact radically improve the current content. I am going to institute his draft changes. I will also save a copy of the older draft and link to it here so that we can integrate facts from it into the new draft if needed and so that other editors interested in this discussion can review this decision. Ill update here in a moment with further info. Jay Dubya (talk) 16:32, 27 March 2014 (UTC)

Something else worth pointing out. Prior to this update, the article relied almost completely on non-English sources. Jay Dubya (talk) 17:15, 27 March 2014 (UTC)

Hey Jay Dubya, thanks so much! And like I said above, totally happy for folks to continue working on improving the article. I've added a "connected contributor" flag here so that there's a record of my contributions to this article. Feel free to reach out if I can help with anything! Cheers, ChrisPond (Talk · COI) 17:46, 27 March 2014 (UTC)

To be clear, I didn't find any serious problems with the replacement draft, my main concern was that I may not have enough background knowledge to detect more subtle, but potentially serious, problems. Gigs (talk) 18:23, 28 March 2014 (UTC)

Gigs - thanks for the clarification. I noted my work background a bit in the hopes that my review based on my technical background and your review based on your extensive wikipedia and editing background would be complementary. I gave another read-through looking out for the sorts of subtleties you note here. Qualcomm is mentioned 5 times in the article. 3 of those they are mentioned among a list of competitors. Twice they are mentioned alone as part of a network-building partnership, however in those businesses a competitor is mentioned in the sentence immediately preceding or following that mention. That's all stuff you likely noticed yourself. The content itself is largely non-technical, focusing more on the history of the protocol and its deployment than the technology itself. This is the biggest weakness of the article as I see it, and is in dire need of expansion. To do so will take some time; we need diagrams and references to flesh it out. Im working on it. In summary of the technological details I feel are at issue, LTE is a protocol that Qualcomm did not create; it was approved by a consortium that included Qualcomm. Qualcomm does not have a market-leading or monopoly position in terms of licensing or deployment of the technology. The article does not appear to be manipulated to twist those facts, or to try and say that LTE depends on some separate technology that Qualcomm did in fact invent or holds a large market share on. I hope that helps explan my review a bit. If you or future editors have any additional questions or concerns that I might be able to help with stop by my talk page. I do plan on adding quite a bit of additional content here once I have some additional references and some art to throw in. Cheers. Jay Dubya (talk) 12:51, 29 March 2014 (UTC)