FamilySearch Wiki talk:The Un-Portal Page

From FamilySearch Wiki
Jump to navigation Jump to search

Problem[edit source]

The subpages located in Portal Pages are not being indexed by search engines, and also complicate editing for contributors. nixiao 14:22, 17 June 2009 (UTC)

Proposed Solution[edit source]

Using a similar layout, pages may be created without the boxes and subpages that currently reside in Portal Pages. For an example of such a page, known as an Un-Portal page, please see United States. If as a community we determine that Portal pages will be moved to Un-Portal pages, we will need to migrate the content from the existing pages into newly created Un-Portal pages. See instructions on how to migrate a Portal to an Un-Portal page. You are encouraged to use one of our many Sandboxes to practice this process before weighing in on this decision.

Deadline[edit source]

We need your feedback to determine how to proceed, but this discussion will close on 3 July 2009 at 3:00 PM, and the final decision will be added to the Manual of Style.nixiao 14:22, 17 June 2009 (UTC)

Discussion[edit source]

Lerman[edit source]

Having dealt with search engines and looking at the source for a portal page, I do not think I agree with the statement that says that search engines do not index portal pages or has problems indexing portal pages.

I looked at the source of one of the portal pages and found all of the "sub-pages" within DIV tags, with specific STYLE, etc. They were not hidden within includes, imports, frames, JavaScript, etc. which would make it more difficult for a search engine to find. In other words, each displayed page is assembled on the server before the user or search engine even sees it.

So, if search engines are having problems indexing a page, my experience tells me that the search engine has not got around to re-indexing that particular page on this particular site. With the above knowledge, it does not make any logical sense to me that going to the "un-portal" style of article will cause anything to be re-indexed any faster.

The above does not mean that I am against the idea. I just want to make sure we make the decision based upon all of the information. It is possible that I am missing information as well. For me, the real questions are whether the portal/un-portal question make it easier, the same, or harder for these categories of people (in this order). It is possible that they may be broken into sub-categories ranging from very experienced users to newbies.

  1. The user that is trying to find information for researching their family.
  2. The contributor that is just wanting to edit a bit of information.
  3. The contributor that is trying to create a page.

It seems that a completed portal page and an un-portal page should look and act nearly identical. If this is the case, the experience & use of #1 should be identical which would be a moot point. This would leave the real questions on #2 & #3.

I will be lazy for a moment by lumping #2 & #3 into the same category for now. Below are some of the advantages/disadvantages that come to mind at this moment:

  1. Contributors have the freedom to completely rearrange the un-portal page and/or make it look completely different. This may be an advantage, but also could be a disadvantage.
  2. I do not remember off hand how one aspect of the portal pages works. My question here is . . . Are the sub-pages always included in its entirety?
    • If so, then it would seem that the un-portal may be easier as one can edit it in place a bit easier instead of having to go to a separate page to edit and it is somewhat out of context.
    • If not, then the portal may be a little easier as it truncates the page to fit.

Well, that is all for now. I cannot think of anything else at this moment, maybe because I have to run to another appointment. Thomas Lerman 16:02, 17 June 2009 (UTC)

Adkins[edit source]

Portal or no, the format of it is very important.

The current US un-portal and the previous US portal both drop very important links out of site on the first page. Those links are to topics for the country. The most important links to at least start on the first screen (without scrolling) are:

1) Getting started, research process, or something of that ilk.

2) States (or counties for a state un-portal)

3) Topics

News, if long, should be a teaser at the most, then a link. It should never be regarded as more important to display than research items, as above.

G Fröberg Morris[edit source]

I appreciate Lermans insight and thoughts into this and agree with his comments. I'm not convinced the UnPortal really does help with search engine indexing. Given the sheer size of 2 million plus articles in the English Wikipedia and their heavy use of the Portal, they don't seem to be worried about Portal or UnPortal. As far as my experience, generally the Denmark and Sweden content comes up pretty quick using key words in a search engine. Maybe the general standardization of titles and formatting has helped. I do like the UnPortal in relation to ease of use. I'd like to suggest leaving the Portals as a Gate, and using UnPortals for SubPortal, or even individual article pages.

Wikipedia is very much "Un-Portal", no boxes to clutter except for info-boxes usually containing basic facts. I've been editing some of the pages there and have yet to see a "Portal" type page. dsammy 08:01, 3 July 2009 (UTC)
I do not know what the percentage is like, but Wikipedia definitely does use portal pages. Here are a few examples: Zimbabwe, Norway, Singapore, Istanbul, Utah, and United States. Yes, I realize that they many (or maybe all) of these pages may have regular articles in addition to portals. My point is just that they do exist. Thomas Lerman 14:56, 3 July 2009 (UTC)
Thomas, if you check the history, there are little contributions compared to the ones not in portals. Knock portals out of urls, you get the much much more information and the public don't even know they exist. Utah portal in Wikipedia is a joke, little information and fewer links and oversized boxes. Again, the public users do NOT even see those portals! The unportal Utah has far more edits and more information. The key is the ease of editing and updating. The portals give me quite an headache trying to edit and update. dsammy 16:39, 3 July 2009 (UTC)
My point is purely that they do exist, are used, and can be found within a search. I am not against nor for portals, just trying to find compelling reasons one way or another and stay objective with the facts. Some have expressed headache both ways. Thomas Lerman 16:57, 3 July 2009 (UTC)
Granted both are complicated to edit, but the portal far more complicated and I tried once and got it so messed up, had to ask sysop for assistance. Portals are scaring off the potential user/contributors. We aren't attracting contributors like we should to those ones in desperate need of contributions. Therefore, ditch the portal format in favor of the non-portal format. dsammy 18:46, 3 July 2009 (UTC)
Oh by the way, in portals, the TOCs are uncreateable. dsammy 18:47, 3 July 2009 (UTC)
I agree that we are not attracting contributors like we should. However, I personally do not believe the portals would be even a major reason. I will stop now as I want to try to keep this to the subject at hand. Thomas Lerman 19:03, 3 July 2009 (UTC)
Thomas, you need to step aside from being a engineer and try simply being a researcher. They don't think the way you do. They think more of subjects that are not being picked up even by Google because they are inside portal boxes. This is the weakness of the Portals. Go to the "Un-Portal" portal format, and they will be picked up more easily. This is my last comment. dsammy 19:12, 3 July 2009 (UTC)
I thank you for you comments. I do quite well looking at problems as an engineer, user, researcher, and from many different perspectives . . . even at the same time. I have seen no evidence that Google does not index things because they are in portal boxes. When some others have looked at the data from a different perspective, they have agreed. I just want to make sure this is an informed decision based upon evidence rather than guesses on how things work. I am not saying that anyone in particular is guessing at anything, just I have not seen any evidence cited. Like I said, I am not currently for or against the change, just trying to collect information to make the informed decision. Thomas Lerman 19:32, 3 July 2009 (UTC) P.S. I found a Wikipedia page on External search engines that discusses this topic, but I never see portals discussed as a problem.
This one is old news. Being a Google account holder, I incorporated Google search engine in my SAMPUBCO site. the results is almost instantly... and updates I make to my site, they show up within 24 hours. So the key is incorporation of Google, Yahoo and other search engines will get faster and sooner results. dsammy 20:02, 3 July 2009 (UTC)
I too am a Google, Yahoo, etc. account holder (for more than one site). That information is not all that old. The key is that every site is different and how often each page gets index varies within a site as well. I have found no evidence nor been shown any that says that portal pages get indexed less frequently than articles. As that article, as well as Google's and other CEO sources, discusses it depends on many things such as the page ranking, how many external links there are and where they are from, how many internal links, site map, robots.txt, various meta tags, how many pages are on the domain, how old the site is, how often the pages change, and many other factors. Any combination of these can dramatically affect how often any particular page gets re-indexed. For example, if all other things were exactly equal (very unlikely), a site with 1,050 pages (or whatever numbers one may want to use) would be fully checked much more frequently than one with 21,400. A search engine does not know how a page was assembled on a server to know whether it was an article or a portal, but the use of frames or some kind of client-side inclues can cause it to not be indexed. Anyone have any source citations on the theory that portal pages do not get indexed as frequently? Thomas Lerman 20:14, 3 July 2009 (UTC)

Maness[edit source]

If going un-portal drops links, please don't do it!  It doesn't sound like there has been enough study or playing around with both ways to truly decide?  Moriss' idea of leaving the Portals as a Gate and having things behind the gate go UnPortal or SubPortal, if it truly will insure they are picked up by search engines ,sounds like an excellent combination of the ideas presented thus far.  

Lerman (links)[edit source]

You are correct, as long as I understand what you were saying . . . links should not be broken no matter which direction we go. Obviously, all links generally cannot be fixed at the exact same time as an article name or namespace is changed. Also, it is reasonably likely that something will be missed. A major difficulty will especially be the current pseudo "breadcrumb trail" style of navigation that many pages use at this time. It has links on MANY, MANY pages. I believe someone was going to check into an automated way of handling these, but could be mistaken.

By the way, I am not sure how I link the way the discussion is broken up here. It seems like it should be more by thread rather than by person. Maybe that was not the intent, but how it seems to have gone and I am propagating the wrong method. I know my initial comments covered several topics.

Anderson[edit source]

Please leave the Portal.  It works well and I do not feel we should change what is working well.

Take a look - Pembrokeshire - the UnPortal and Pembrokeshire the Portal. That is the difference. dsammy 21:55, 2 July 2009 (UTC)
Hello Anderson, thank you for stating your opinion. I have not heard nor seen any compelling reason one way or another. Sure, some portal pages may be underdeveloped or look bad, but I am sure some non-portal pages can be in the same condition. Thomas Lerman 22:12, 2 July 2009 (UTC)

Samuelsen[edit source]

I am all for the UN-portal. Been waiting for the changeover since May 5 or was it 13th, day before the NGS conference. Lost links are easy to track down than you think and added back on the UnPortal page. Bread crumb trail is not a big deal when done right.

Seoane[edit source]

I know that the deadline has past but I just came back from vacation and learned about this discusion. just wanted to put y two cents in. From the conversation it seams that both the Un-Portaled pages are searchable and the current Portal pages are searchable. The question from a consumer standpoint that should be answered is:

What is the easiest format for me to contribute in?

If the question has been answered as to searchability then why not change over to the most patron friendly format. From the discussion it looks like the Un-Portal pages are the eisiest to use.

Do we want a greater amount of individuals to contribute or not?

I agree . . . if the "un-portal" pages are truly easier to create and edit, then we should use that instead. I personally do not feel like we have heard much response to which is easier or not. Yes, we want to get a much greater number of people to contribute and therefore the easier the better. Thomas Lerman 15:58, 6 July 2009 (UTC)

Dilts[edit source]

I vote for un-Portal. I like the ease, simplicity, flexibility and cleaner look. A user can find plenty on a regular page, and learn to edit it faster in my opinion.

Whether a portal is easier for search engines to find and list I know not. But I do know trying to create a portal page is much more complex than creating a nice regular page. Diltsgd 17:46, 7 July 2009 (UTC)

[edit source]

Once again, it is past deadline but I think everyone is overlooking some major points about Portals. Portals are NOT meant to store/present data or information, they are meant to show a condensed set of links to allow you to find a specific sub-topic, and basically make a title page or table of contents just as if you were to make a mini-wiki just for that topic. Several users heretofore have mentioned how Portal pages have significantly fewer edits than "Un-Portal" pages, placing blame on the fact that they are harder to edit, but that is not necessarily the case. Portal pages in Wikipedia rarely are edited because once they have gotten to a certain point of being set up, and all the commonly used links are in place, the layout is set up, and there is no reason to bother them.
I would like to further this by pointing out that in almost all cases in Wikipedia, both a Portal and a Mainpage exist because they both serve their own separate purpose: a Portal to show and link available resources, a Mainpage to show information/data/facts, and Sub-Pages to go further in-depth where appropriate. I think it would be useful to follow the pattern established by Wikipedia and use them both.
Now arises a new question, where is it that we primarily want to direct our users, the portal page, or the main article? I guess that lies in what will make it easier for the general user-base to find and get what they are looking for. Currently, being tucked away in beta, the general user-base is a bunch of nerds (I don't mean that pejoratively) who have found it and are trying to get it going, but in the long run we want to cater it to a much wider audience.
In my wanderings in Family History work, generally speaking, there are two groups of people:

  1. Those who are just starting and aren't familiar with research methods or sources.
  2. Those who are more or less experts.

I guess the way I see it is this:
Beginners -> Main Articles: Those who are trying to figure stuff out will want to see more text. Start off reading something to get their bearings and try to figure out how to find an ancestor. An example thought process of one of these users would be something like this: "So... it says here that (x information) was recorded in (y location) so I can find my ancestors that way".
Experienced Users -> Portals: Usually these users have and idea of what they are specifically looking for. They would rather not have to scroll through a large article trying to find the link that they are looking for. An example case would be something like: "Ok so I know he died in (x location) so I wonder what newspapers would have published his obituary?" The user would then desire to jump from the State to the County to a List of Locally Published Periodicals. This user would ideally want to get to this information with as little scrolling and as few clicks as is possible.
Obviously we want the wiki to be accessible to both user types. I think in the long run it would be best to use Main articles as the primary target for links, and keeping portals kind of like a hidden back passageway. As users learn and become more experienced in researching they'll discover how portals work and "graduate" to them.
Jacob.creedon 19:28, 10 July 2009 (UTC)

Anne Wuehler[edit source]

I vote to remove the portal pages. I teach an introduction class in the Family History Library about the Wiki. Part of the presentation includes how to search. I have a little problem trying to explain why two entries appear in the result list for the same locality, one starting with the word Portal and the other not. Anne 17:17, 18 September 2009 (UTC)