The following discussion is an archived debate of the proposed deletion of the miscellaneous page below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the page's talk page or in a deletion review). No further edits should be made to this page.
The result of the discussion was: keep was never really a delete nomination, so not surprising there is no consensus to delete. Excellent discussion, all. (non-admin closure)UnitedStatesian (talk) 16:16, 29 April 2019 (UTC)[reply]
Long post: is this worth keeping? Nomination without recommendation, to trigger a discussion.
I am Irish. Most of my major interests relate to Ireland. So in theory, if we're going to have more than a handful of portals on Wikipedia, I'd like to see a portal for Ireland.
But in practice, this portal has never risen above the so-so level. The problem is not with their efforts; it's with the whole way in which portals are implemented, and possibly with the very concept. Per WP:PORTAL, "Portals serve as enhanced 'Main Pages' for specific broad subjects", but in practice they do not add much value. Portal:Ireland as it is right now is less an "enhanced main page" and more of a slim single-issue magazine.
It currently consists of:
The lede of the head article Ireland (or at least the three introductory paragraphs thereof)
A selected picture, currently of kids playing hurling in ummm ... Philapdelphia
One selected article, currently Áras an Uachtaráin, displayed as lede plus image plus click-to-see-full-article
One selected biography, currently Pádraig Pearse, also displayed as lede plus image plus click-to-see-full-article
A "topics" section, which is actually just the navbox ((Ireland topics))
A "Categories" section, which is just a list of the first layer subcats of Category:Ireland
A list of Featured articles
A "Did you know" section, with 3 mediocre hooks
The infobox of a selected city
Refresh, get another single issue .. and that's it. I know that it has taken a lot of work to get it this far, but the result is underwhelming.
I see little here which fits the sort of criteria described at web portal. No multiple sources (apart from DYK), no dashboard, no aggregation. No gateway to other stuff. The only really useful bit is the list of featured articles.
Anyone looking for some sort of overview of the major topics won't find it here. There is no list of counties, of major landforms, of dominant sports, of major biographies by field or era, of landmarks in history or politics, no mention of economics, no section on the arts or architecture.
What we really have here looks more like a slim single issue of an inflight magazine. Those consist of snippets of gloss produced by young journalists hoping for a job elsewhere in real journalism, all with the aim of boosting the company and selling advertising space in a publication which survives because of its captive audience. The constraints of format mean that the snippets displayed here are inevitably a pale shadow of the scholarship and fine writing which ww2censor, Tóraí and others produce elsewhere on Wikipedia.
Wikipedia portals don't have a captive audience. Our readers our not buckled into a chair for an hour with an inflight mag, an overpriced drink and a dream of plastic-trayed food with plastic cutlery.
Readers simply don't want this "slim magazine" format. The table on the rights show the daily average pageviews for January–February in each of the last few years. The results are similar to those for almost every other portal, apart from the 8 listed on the top right of the main page. Across a huge range of topics major and minor, there is a steady pattern that readers prefer the head article by a ratio of between 100:1 and 2,000:1.
So what do we do?
Despite all the work put into it, it seems to me that the current portal isn't really worth keeping.
It could be updated and expanded, but for a decade no editor has sustained that for long. It doubt that will change.
The one-snippet-at-a-time layout could be upgraded to the Lua-based slideshow format of the automated portals. But that format seems to me to be is pointless. An alternative view of a single page is simply redundant to the default view of a lists of links, where mouseover on any of the linked list items shows you the picture and the start of the lede. This is hidden from logged-in editors who use WP:POPUPS to provide an editing/analysis menu in place of the preview, but most of our readers (for whom we create this stuff) are not logged in. Using the default view of Wikipedia, the bare list of bears actually gives readers much the same functionality as the so-called "portal" view, but without all the overhead of a "portal". I have tested this in Opera, Firefox, Google Chrome, and MS Edge; it works on all of them. Try it yourself using the "open in private window" function on your browser.
So I reckon we would do better to dispense with both the Lua code and the "selected article" subpages, and have something more like a compact version of e.g. Portal:Contents/History and events, which is just a lot of bare links. That would still display previews, just via mouseover; and it would provide a much better portal to the vast range of Wikipedia content on Ireland. Instead of being magazine, it would mirror the multiple links structure used by for example the award-winning https://www.gov.uk/ website.
However, that would be a lot of work to create, and even more work to maintain ... and I don't see much evidence that readers would actually want it. Even the Portal:Contents/History and events I mentioned above got only 172 pageviews per day in January–February 2019, which is less than 5 times the views for the current Portal:Ireland, and only still 2% of the pageviews of the article Ireland. In the grand scheme of Wikipedia topics, Ireland is a much narrower topic, so we'd get even fewer views than that 172.
So I can't see that any of these paths is worth pursing.
And I remain unpersuaded that any possible portal would actually get much use, even if a massive effort was put into it. Two reasons:
Massive hyperlinking. Wikipedia pages are so heavily interlinked that even a modestly well-written head article on a topic is of itself a portal. This isn't like the mid-1990s web, when web pages were mostly plain text with a few links at the top and the bottom; rich interlinking is now the norm, and it seems to me that portals are redundant.
Search. As web analysts such as Jakob Nielsen noted as early as 1998, good search killed navigation, because users found it much easier to search than to navigate a website's menu structures. That's why search suddenly became de rigeur on web sites, and why the major web portals such as Yahoo fell off a cliff. Readers simply didn't need portals any more; they became like road atlases in the era of satnav.
But what do we do with Portal:Ireland? Keep the current inflght magazine, as it grows ever more out-of-date while readers shun it? Put a major effort into a complete redesign and rebuild, pray for ongoing maintenance, and hope that readers will follow? Or should we just say that we don't see any way of devoting enough effort for a major overhaul, don't see any model of Wikipedia portal which readers actually want ... and delete this one without prejudice to what the community does with the rest of the 539 portals? Any other suggestions?
I wondered whether this was best done as a post at WT:IE, but I think it will get a wider audience here. But I do hope that we can have an actual discussion about the options, rather than simply one-line keep/delete !votes.
I am much less interested in the outcome than in exploring the options. --BrownHairedGirl(talk) • (contribs) 19:33, 21 April 2019 (UTC)[reply]
OK, first in, but I'm in the right mood, after a couple of weekend days off. To take it off the table, I favour deletion. But as BHG notes, voting is not the point, the effort put in "by divers hands," and the nature of the topic, merit debate. Not to sound like an echo, but I tend to agree that the time for most portals has simply passed. Just as many of us no longer bother keeping long lists of favourite articles, but just use a tool like Google, so the era of static portals came and went, actually at least 5 years ago, only we did not notice sufficiently, or act. I started in IT when Gopher was fresh and webpages were often tiny and indeed linked top and bottom in chains - but this is prehistory. Nowadays people rely on rich and deep wiki-linking, and search, and all else is a little redundant. And the viewing figures show this. Readers arrive to the main topic pages. It's just the current reality. And this takes me to the kicker - and a point I've underlined in other debates. We only have so much community resource, so much editing capacity. Not enough - we all know we need more editors and admins - but we survive. But, and sorry for the manager inside coming out - we need, as far as one can in our free-wheeling community, to focus the capacity we have. And portals are just not "good value for editing time." I'm in the middle of a 16000 article review quest, and I'd *far* rather have more hands to improve many of the 7k reviewed to date, than working on a portal almost no one visits. This issue is acute for an area like Ireland, with a modest few dozen regular general editors, and no more than 100 (my guesstimate) even counting specialists. Now, could we just leave the Portal? Yes, but it does a poor job for Wikipedia. SO better to prune, unless someone is clearly willing to step up and take a lead role for a time. My tuppence worth...SeoR (talk) 20:48, 21 April 2019 (UTC)[reply]
Mnay thanks, @SeoR. That's well-informed and thoughtful. It's a bonus that I agree with it all of it, but the most important bit for me is that you have a sense of good of how the web and en.wp have developed.
And yes, I remember Gopher. 14 years after I'd first been a sysop on a PDP-11, it felt like the first real user-friendly info-sharing tool. For a few months I wondered if this new Web thing with its bloated flaky interfaces would ever be more than a fad like Apple's HyperCard, but it soon became the future, and after a few years of competing navigatonal models, Google search basically defined the navigational system of the web we have today. Essentiallly, if you want it and there isn't a link in front of you, just search. I'm not usually a great fan of the WMF's technical priorities, but they have fairly good job in making Wikimedia's search function effective for the varied needs of content editors, process wonks and readers.
So, yes I agree that this portal would be better pruned. But I'm a little wary of the middle manager logic. Wikipedia editors are volunteers, who devote their time to tasks which they like doing, not to those whatever they have been directed in return for a pay cheque. So if those who like making portals don't get to make portals, they may not just shrug and turn their hand to reviewing or to writing featured articles. OTOH, it probably would be good not to have this hanging around as an unguarded black hole to suck the energies of an editor whose attention could otherwise go to real content.
However, I'm puzzled by your final comment unless someone is clearly willing to step up and take a lead role for a time. Support some smart, skilled energetic editor appeared. Someone with good skills both in content creation and in informatics who said yes, they'd love to put in lots of time over the next 6 months to improve Portal:Ireland. What would you hope they they would do to it?
I think this is an important question, because there are a small group of editors who clearly do want to keep portals ... so what should our eager volunteer be trying to create for them? --BrownHairedGirl(talk) • (contribs) 21:57, 21 April 2019 (UTC)[reply]
@SeoR: I disagree radically with most of this, but just quickly taking one point: deleting a portal on which one or a group of people have worked will not make those people edit articles, it will make them leave Wikipedia. I can only speak for myself, but for me, the portals I help to maintain (including the main page) are my way of triaging the nearly 6 million articles on the encyclopedia for attention. By that I mean that I copy edit & source articles linked from the main page; add news items from daily browsing BBC news & where appropriate edit the article; look for new content to include by running through categories & Wikiproject-tagged articles, editing them as I go; update my portal excerpts & copy edit/correct/update the main article at the same time; and create articles with an eye to DYKs for the portal. Espresso Addict (talk) 22:53, 21 April 2019 (UTC)[reply]
I hope it's OK if I cut in a reply here. So both @Espresso Addict: and @BrownHairedGirl: make the point that some editors - and yes, we're all volunteers, and no one is ever going to "manage" WP (at the first sight of such, many would just quit, I know) - like to work on portals, and will leave if they're not available. I don't disagree, and I did not !vote for "delete all portals," ever. But the question is about a specific portal or portals, and for these, some firmness is needed. If the portal for an entire actively edited country is little visited, and less maintained, I do believe it would be better not there at all. There will be suitable pages elsewhere. If someone were going to actively "work the land" of Portal Ireland, they'd be doing so by now, but in a sample set of many years, they have not. How long should we wait, leaving something weak in place?
So, to the second question from BHG. It relates to the above. I'm not opposed to the concept of a meta-article, call it a Portal. But such a thing can only be valuable if it truly goes far beyond what a well-linked article and category tree can do. So this point was a little bit the compromise part of me at work - I would not like to say there could never be a Portal for Ireland, but it would need a new vision, and a very strong person or dedicated team to make it meaningful. I would not go for that myself, but I'd never stop someone else. But again, the evidence of years is that this probably is not happening, and I believe the points I made previously indicate that the chances are low, and the demand from the masses of readers globally is simply, basically, not there.SeoR (talk) 07:33, 22 April 2019 (UTC)[reply]
Great example - I have looked at 2 snippets. Last modif: 2008 (except from pictures-robots). The baby seems dead, but I will use the famous WP:NOTCOMPULSORY to avoid looking further. A great case of Let the dead bury the dead[1]. Pldx1 (talk) 20:50, 21 April 2019 (UTC)[reply]
Keep – Meets WP:POG per topical scope, breadth and content availability. There is also a fine amount of Recognized content, in the form of Featured articles. Also, sorry, but the nomination is partly erroneous (e.g. "One selected article", "One selected biography"):
There are presently 45 Selected articles. See Portal:Ireland/Selected article archive. These can be viewed by purging the page or selecting the "Archive" link. For whatever reason, a purge page link was not present on the page prior to my reversion of portal automation, so I added one.
There are many more DYK entriesavailable at Portal:Ireland/Did you know. These can be made to randomly appear on the portal by using a ((Numbered subpages)) template and integrating the list into separate subpages, say three or four DYKS per each page. If the portal is retained, I may consider performing this.
More Selected pictures are available at Portal:Ireland/Picture. These can also be utilized using Numbered subpages
– It's very important to check for the presence of content on portal WP:SUBPAGEs when assessing portals for potential deletion. Many portal updates occur on the subpages, rather than on the main portal page. North America1000 21:27, 21 April 2019 (UTC)[reply]
Addendum – I have fixed the Selected series: Irish cities section of the portal by creating a Portal:Ireland/Random City page using the ((Numbered subpages)) template. Now 11 entries are shuffled when purging the page. North America1000 22:04, 21 April 2019 (UTC)[reply]
Congratulations, @Northamerica1000, for completely ignoring the nomination, and completely ignoring the hope that we could have an actual discussion rather than bullet-pointing factoids. --BrownHairedGirl(talk) • (contribs) 21:47, 21 April 2019 (UTC)[reply]
Well, you stated in the nom that there are only single entries available, further stating, "And that's it". This is incorrect. Why not address the errors in your nomination? North America1000 21:54, 21 April 2019 (UTC)[reply]
@Northamerica1000, obviously, refresh, get another issue. I guess I should have stated that explicitly, so I have just added it. But it's still just multiple editions of a slim magazine.
You have completed ignored my wider questions about what use this is when hardly anyone reads it ... and what a portal might look like if we set it to make one which readers might actually read ... and whether anyone anywhere on Wikipedia has produced a model of portal which readers actually use.
This is what has saddened me so much about the portals project's response for the last year to the whole portals debate. It has all been about creating X portals with y links etc, and "this topic deserves a portal and here's a page called a portal" ... while studiously avoiding the fact that so far none of these designs attracts more than a tiny fraction of the readers who are interested in any given topic.
Please can look up from the details and try to think about the big picture? --BrownHairedGirl(talk) • (contribs) 22:09, 21 April 2019 (UTC)[reply]
@BrownHairedGirl: I've seen a trend in a lot of these discussions where editors seem to equate a lack of page views to lack of interest in a portal (on a particular topic, or just in general). I would submit that that is not necessarily the case, with the most likely reason for the low page views being that the majority of readers aren't even aware that the portal exists. Consider, the only place deemed acceptable to "advertise" a portal is a (rather easy to miss) ((Portal)) template at the bottom of the page - a place where many readers might not ever reach in the first place (sans following a citation). This is especially pertinent considering the ever-growing mobile readership. It would only seem natural that a portal would claim only a small fraction of it's corresponding article's pageviews in light of this, and was one of the reasons I asked for more data on this exact topic.
Far from being an indicator of reader disinterest, this seems to instead show a failing of the portal system to properly advertise itself to readers, who might otherwise be interested, had they known of it's existence. Also, I wouldn't really expect pageviews to correlate to reader interest, even if the numbers were more equal between the namespaces. We only have one point of data to work with: the user clicking on a link. That doesn't really tell us of their experience beyond that point. Did they stick around and read the portal/article? Did they find what they were looking for? Did they click through to another topic that interested them? Did they immediately close the tab? That type of data isn't possible to glean from pageviews alone, which is why I consider them a poor proxy for estimations of reader interest.
Short of doing A/B tests or more in-depth analytics, we are unlikely to get that type of information either. A pity, since better data would really help us objectively gauge the situation and decide on a path forward. This is slightly tangential to the current discussion, just my thoughts on the matter. — AfroThundr (u · t · c) 00:01, 24 April 2019 (UTC)[reply]
@AfroThundr3007730, I hear that "more promotion" argument a lot, and it would obviously be good to have some proper research n it.
However, the evidence I have seen so far leaves me unpersuaded that anything short of advertising-style hog-the-limelight promotion would make much of a difference. As it happens, Portal:Ireland is exceptionally well-linked: 3,061 links from articles, 1173 from categories, and 8 from portals. That's pretty good going, but the portal still gets pitiful pageviews: only 38 per day.
So it seems to me that on a pageviews-per-link ratio, the portal is underperforming by a factor of about ten.
The one type of experiment I have thought of so far that wouldn't be too spammy is to link from infoboxes to portals. Maybe something down at the bottom of the infobox, near where website links go. --BrownHairedGirl(talk) • (contribs) 01:09, 24 April 2019 (UTC)[reply]
Formal keep, for the record. It's difficult to have a discussion when the page title is "X for deletion". This is a well-populated former featured portal on a topic that I hope BrownHairedGirl & I agree is broad, interesting & important. People seem to have widely differing opinions as to what the consensus of the RfC last year was, but I don't think it was to delete well-formed, longstanding portals. I'm happy to have a debate with BrownHairedGirl at a more appropriate forum (indeed I meant to start one the other day but got drowned in portal MfDs) as to where the community should go with undermaintained portals in general, now that the widely advertised MfDs started by her & others have shown that The Transhumanist's reboots with a template-based autoportal are not supported. In summary, I subscribe to the magazine model of a portal, though I hope my best one is better than an inflight magazine. The UK government website is an entirely different case: nearly all visitors to the site will be wanting very specific information (how do I register a death, what are my responsibilities as a landlord...) rather than browsing for interesting content on the UK system of government. But this is not the place to have that discussion. Espresso Addict (talk) 22:20, 21 April 2019 (UTC)[reply]
Thanks for your comment, @Espresso Addict. As noted at the bottom of the nomination, I did consider positing this at WT:IE, but I came here in the hope of attracting more participants, and because MFD has been the only venue where there has actually been substantive about what on earth these portal pages are for. (WT:WPPORT is all nearly all details)
Anyway, I'm interested that you support the magazine model. Why? Is there any evidence anywhere in Wikipedia that readers actually want it? Is there a cluster of portals on major broad topics (i.e. VA-levl-3 or higher) which routinely get say 10% of the pageviews of the head article? Or even 5%? Or is there some external evidence to support this way of displaying content? -BrownHairedGirl(talk) • (contribs) 22:44, 21 April 2019 (UTC)[reply]
I came up with the magazine model a few weeks back in offline discussions. My background is scientific/medical publishing, and I once edited a "magazine", ie a review/opinion/news journal on an academic topic, so portals seem to me like that: a free magazine where all the back issues are available online at once. (People still buy print magazines for money; a penetration of 0.5–1% of all people interested in topic X would be considered pretty decent in the print publishing world.)
As to readership, I don't fully understand this statistic, but take a look at [1] (Cheshire was featured, Derbyshire is an adj county that wasn't): my formerly featured portal's hits decreased ~fivefold literally overnight when the featured process was fully deprecated, without changing in any way except the fill colour of the little star, so I don't think low hits (or for that matter high hits) are necessarily related to the quality of the portal. From the portal-philic perspective, I'd like to recast the let's delete portals because they don't get hits, into how can we get more hits for (good, maintained) portals. But I've had no luck getting anyone to debate that one.
And now I really must do go & something more useful than debating portals... Espresso Addict (talk) 23:22, 21 April 2019 (UTC)[reply]
Interesting, @Espresso Addict. I can see your interest in magazines, but comparisons of reach between free webpages and actual physical purchases are quite strongly apple-and-oranges.
But it strikes me that if your goal is to create a set of online magazines, the you need a new namespace, like "Book". I have yet to see any sources suggesting that a magazine is a type of portal, it feels like a bit of a hijack to try to turn the navigation-and-summary-based concept of a portal into a vehicle for magazine-making. --BrownHairedGirl(talk) • (contribs) 01:15, 22 April 2019 (UTC)[reply]
I have no interest in looking at the current portal guidelines, which have been edited back & forth so much over the past year or so that the history makes my head spin, but that used to be what portals were about, at least in my understanding. I think one reason we are all talking past each other so badly is that different people with different backgrounds perceive them differently. (With a print publishing background I see them one way, but others with a computing background see them entirely differently.) I see the plain-link version you talk about as essentially an outline or directory. I have no objection to these (when well done) but no interest in developing them, either. I see no reason why the 'pedia can't have both. And navboxes, categories, search engines... IDIC. And in the time I've taken debating the existence/importance/function of portals in the past month or so I could have done a lot of other more interesting/useful things. This discussion has no purpose in the sense that nothing that comes from it can be binding, except the deletion/not of a portal that even you, as nominator, don't advocate for deleting.
If you seriously want my advice, BrownHairedGirl, I'd suggest that you start/develop a portal about a topic you care about & think is sufficiently important, and then come back when you've got something you're moderately happy with and ask portal experts for peer review. I'd be happy to look, and who knows we might both learn something useful. Espresso Addict (talk) 01:37, 22 April 2019 (UTC)[reply]
One further comment, if you want an example of a portal that gets as many hits as its head article, you could look at the main page, which usually gets a lot more hits than the TfA: [2]Espresso Addict (talk) 01:49, 22 April 2019 (UTC)[reply]
Delete, just to get it out of the way. But now the broad discussion: I think there are some subject areas that are so broad that readers would be served by having some way other than the article to navigate them; so broad that a navbox could not possibly be created to cover them. I believe there is at least a chance that portals could be this way to serve readers: to increase interest, encourage exploration. One can only imagine the good place where we would be if those interested in portals had, a year ago, focused on making one perfect portal, rather than trying to create as many portals as possible. But they didn't, so here we are. Unfortunately that breadth-of-subject-area criterion that creates the need for a portal in the first place leads to uncomfortable discussions and emotion. The "why doesn't the "x" I am interested in get a portal too" is answered with a simple, but still emotionally painful, "because it is not a broad enough subject." So yes, Portal:The Beatles, but sorry, no, Portal:Kelis. Yes, Portal:China, but sorry, no, Portal:Ireland. Yes, Portal:New York City, but no, Portal:Erie. And of course yes, Portal:Mathematics and Portal:Opera.
Where to draw the line? as you know from reading my MfD nominations and !votes, I, like a few others, have found Wikipedia's vital articles helpful, as the participants there are forced to put articles into tiers; interestingly, the recent MfDs that are not TTH related have largely fallen out along level 1-2-3 kept, 4 and below deleted. Only 36 countries make the WP:VA3 cut there; Ireland does not. But whether we use vital articles or another criterion, we cannot always go to the "every "x" gets a portal" (which got us into this mess). Or "no "x" gets a portal."
And broad subject is necessary, but not sufficient: a portal also needs dedicated maintainers: a group that can change, but can't be absent. But note the converse is not true: one, or a group of, dedicated maintainers cannot overcome the bresdth-of-subject-area requirement. UnitedStatesian (talk) 03:05, 22 April 2019 (UTC)[reply]
Personally I think there's a sweet spot on breadth; too wide and it becomes disjointed or feels unrepresentative. Who'd look twice at Portal:Biography, covering all bios under the sun in 12 (presumably) featured articles? Might as well just click on the random article button a few times or pick someone from the list of FAs. I think decent coverage in Wikipedia is critical, perhaps the most critical factor and certainly far more so than the vital article team's efforts; thus English-language topics such as Ireland or Minnesota are more likely to be able to support a portal, and, conversely, the likes of Myanmar & Angola struggle. Espresso Addict (talk) 04:37, 22 April 2019 (UTC)[reply]
But I find it interesting that you can't seem to bring yourself to draw any line (or clarify whether your "decent coverage" is a qualitative or quantitative term): any collection of articles greater that 20 was the default position before (a position that never received broad consensus), and I have seem no one who held that position adopt a different standard. I note there are 19 major league baseball award articles that are FAs; would you say that getting just one more to FA would justify Portal:Major League Baseball awards? UnitedStatesian (talk) 05:52, 22 April 2019 (UTC)[reply]
I mean decent coverage; it's one of those "I know it when I see it" things. A mix of quantity & quality. The smallest project-tagged article set I recall seeing made into a reasonable portal was Portal:Barack Obama, back in 2010 when it was being reviewed for featured portal, when there were as I recall just north of 500 project-tagged articles, but obviously given the editor bias they were of higher than average quality. I can't answer questions relating to US sport, I'm afraid, as a sports-phobic Brit I lack the relevant background. Espresso Addict (talk) 07:46, 22 April 2019 (UTC)[reply]
ETA. amusingly enough, fwiw Ireland seems to have more project-tagged articles than China. Espresso Addict (talk) 07:55, 22 April 2019 (UTC)[reply]
The place of portals in Wikipedia is a legitimate topic for debate. The tactics being used here are totally deplorable and disgraceful. This is not a discussion about one portal. It is transparently one of the later phases in a war of attrition aiming to overturn the massive RfC of a year ago when it was decided to keep portals as a concept. This discussion should be closed now and transferred to a widely advertised RfC stating that it is re-opening the question of portals in general: Bhunacat10 (talk), 19:39, 22 April 2019 (UTC)[reply]
That's a fine rant, @Bhunacat10, and I hope you enjoyed writing it. It's a pity that you didn't read the closing line of the nomination, in which I write that I am much less interested in the outcome than in exploring the options.
This explicitly not an attempt to open WP:PORTALS. It's a discussion about what options are available wrt this particular portal.
I am well aware that wider decisions need to be had. The last few months of drama arises out of the disruptive antics of WP:WPPORT, which set about misrepresenting the decision of WP:ENDPORTALS not to delete every single portal as a licence to nearly quadruple their number by creating over 4,000 new portals, the vast majority of which were simply automated spam. Those who objected to that were simply brushed off as a script-jockey and his accomplices built uncurated garbage at the rate of one every 60–120 seconds and circulated newsletters which boasted solely of the number.
That was a blatant, shameless attempt to create a WP:FAITACCOMPLi. Supporters of the steamrollered spam have fought a sustained, wikilayering war of attrition to resist a cleanup, wikilawyering every available opportunity to make a cleanup as difficult as possible. They opposed speedy deletion, because they wanted the community to spend lots of time scrutinising every one of the pages which was created without care, in seconds. (That's absolutely class gaming the system). They objected to single MFD nomination as too many discussions. They objected to mass nominations as too few discussions. That's the war of attrition.
Now, as the spam cleanup phase nears a close, we need to focus on what to do about remaining portals. Some of that needs to consider broad principles, but we also need to consider how those principles apply to individual portal and to the limited pools of editors working in many topic areas. So this an attempt to start a discussion about what are the actual options for the future of one portal which remains neglected, outdated, almost unused.
It seems that you have nothing of substance to add to this discussion, which is fine. But it's not so fine that you come here to make a personal attack on the basis of your blatant assumption of bad faith and offhand rejection of explanations of intent, and it's not fine that you label as deplorable and disgraceful an attempt to start a discussion. --BrownHairedGirl(talk) • (contribs) 21:40, 22 April 2019 (UTC)[reply]
I could say the same thing to you, BrownHairedGirl. Very little of that is relevant at all to this non-automated portal, which is not authored by The Transhumanist, nor does it use their methodology, and long pre-dates their interest in portals. The problem we have here is that, deliberately or accidentally, you've opened a deletion debate in which several people have already supported deletion. If Portal:Ireland were to be deleted, or even closed as no consensus, that acts as a precedent for deleting other portals, lots of them, in fact I'd say 90% of the geographical ones. It's hard – and I'm good at assuming good faith – to think that you didn't realise this was a possible consequence in advance. We could have had this debate in your user space, or mine, or at the portals project, or at the talk page of WP:Ireland, without the risk of setting any precedent. Espresso Addict (talk) 04:01, 23 April 2019 (UTC)[reply]
@Espresso Addict, I was replying to an editor whose only contributions I have seen to portal MFDs have been outrage that even the portalspam is being deleted, and who came here solely to express insults and further indignation that anyone who would actually hold a discussion. So, my comments about the spam are relevant.
I see one explicit delete vote. And I'll add mine as a "keep for now" to make my intent v clear.
A debate in userspace would just be a private discussion. I want a public one.
A discussion on WT:IE would not have been visible to so many editors.
And as to the portals project ... when I tried last September to have discussions there, it was what I can most politely describe as a madhouse, full of multi-screenful dumps of rage about about how the wizard new technology would do xyz ... and almost entirely evading any discussion of what the pages were for. So it was here or the village pump.
The fact remains that this portal is rotting. It is unmaintained, outdated, and barely used: a zombie portal. I'm asking for solutions, for alternatives to deletion, because I can't see them. We need to have this discussion, because right now there are no answers.
And I think possibly the most important contribution to the whole discussion on this page has been our acknowledgement that 90% of geographical portals face similar issues. Now that most of the spam is gone, we are likely to see actual genuine please-delete nominations of zombie portals for major topics, with sound policy-based rationales. I'd much prefer to at least have one discussion framed around seeking alternatives.--BrownHairedGirl(talk) • (contribs) 04:31, 23 April 2019 (UTC)[reply]
Thanks for the explanation. WP:IE might have been the best venue; you could have advertised it at the portals project, and/or dropped a note to those who have been commenting here these past weeks.
I didn't actually mean to state that 90% of geo portals have similar issues, though it's possible, just that 90% of geo portals are based on a similar or smaller geographical area/article base/population base. You might not see the precedent, but less-careful nominators might. (For example, Portal:BBC is currently nominated and the nominator gives Portal:CERN as a precedent, despite the fact one is what I'm abbreviating as a "one-click wonder", and the other (for all its failings) isn't.) On the quality issue, I ran through quite a lot of the featured portals back in ~2015 with retired editor Sven Manguard, and we found relatively few supposedly featured portals that were entirely free from issues – and that was not checking the lead extracts for currency, except in the broadest (the subject's dead or the article's deleted) fashion. I doubt never-featured portals are any better, and the situation has certainly not improved since then. What tended to happen was editors/projects put work in to get the portal featured, and then just stopped maintaining it altogether. Often linked content that had been GA deteriorated substantially over time, turning into orange-tagged C-class muddles. (This is one of the many reasons I don't necessarily favour transcluding lead extracts; in my experience, FA leads are often stable, but GA/B-class leads tend to deteriorate.) Espresso Addict (talk) 05:12, 23 April 2019 (UTC)[reply]
As explained, I posted here to get a wide audience (and hopefully participation).
Look, here's what I think is going to happen. Either we get some sort of solution to the sort of issues which I raised here, or there will be ongoing chiselling away at the collection of substandard portals. Some of that may be done crudely (as with poor bundling), and some of it may be more carefully reasoned and more successful. But after the portal project destroyed its credibility so comprehensively in the last year, deletion proposals will in general have a fair wind. There is nothing to stop the portal project now having a discussion on what to do deal with the systemic problems in a way that will persuade the wide community that more portals worth keeping ... but I doubt that it is really able to shift to actually listening to community concerns rather than shouting down and wikilawyering. I'd be happy to be proven wrong, but so far as I can see, for every reasonable portal fans like yourself, there is at least one like editor I was replying to here who is till either seething that TTH's spam lake is being drained, or that reverting to broken manual versions is a solution.
I see your point about transcluding lead extracts, but I think our solution is completely wrong. The whole business of maintaining forks of ledes is just systematised WP:REDUNDANTFORKs, leading to them becoming increasingly outdated as you describe. Eliminating that is one of the things that TTH got right. IF former GA leads are deteriorating into orange-tagged junk, then we should fix those ledes rather than leaving them to rot 'cos the fork with 1% of the pageviews is still OK.
It wouldn't be a huge job for someone with tools to select all the /selected-article subpages and mass MFD a subset of e.g. those over a certain age about BLPs as broken redundantforks. It would be hard to dispute the problem. --BrownHairedGirl(talk) • (contribs) 06:14, 23 April 2019 (UTC)[reply]
I ended up taking a long wikibreak last year; I think a lot of the more reasonable members just left. Most of the wikilawyering over there seems to have calmed down a bit, but it's hard to get a discussion that goes anywhere; there are some fundamental disagreements and a lot of talking past one another. To be honest, I was hoping ArbCom would take the thing on, even though I know that never ends well.
Obviously the ideal is that portal extracts get used to repair degraded leads. I've done this myself, but one sometimes gets pushback. It's been my experience, for example, that some members of the medical project don't much care for major edits of "their" articles by outsiders. And there's all the problems with focus, length, style/language... Not as trivial as it seems.
I systematically check all my BLPs fairly often (at least every few months when I'm active), but don't edit the excerpts unless there's a significant change; the ones about long-retired people rarely have significant updates but one or two of the subjects have died on me. (I'm a bit paranoid about one of them dying without my noticing.) So some might look abandoned but not be. It's quite a lot of work just keeping up with all those on two small manual portals, which is one reason why I'm resisting pressure exerted here to take any more portals on.
It's a shame one can't set up an alert for subject died; thinking about it, with Wikidata that should be technically possible. I asked at the portals project about the possibility of using Wikidata somehow to update things like annual polio figures, but got no positive response.
In the general case, I'm not seeing any alternative to hard manual work vs going for transclusion. One of the things that I particularly disliked about last year's events was the fact that hard work on portals was deprecated. (No-one collects DYKs by hand, it's too difficult!) But, honestly, it's hard to motivate oneself to keep on updating all the non-BLPs, when one feels the portal is under constant threat. Espresso Addict (talk) 08:10, 23 April 2019 (UTC)[reply]
I get that problem of deletion-threat deterring maintenance and improvement, @Espresso Addict. It's real. But it's also an inevitable consequence of a long-term unresolved problem.
We got here because a decade without a Sword of Damocles left us with a pile under-maintained, largely unused pages, and because the project's response to threatened annihilation was to make the problem significantly worse, both in scale (by adding so much spam) and depth (by degrading so many pages). So a "take you time to fix it" approach is hard to justify.
If there was now some sort of plan to address these long-term problems, or even some sort of process to develop a plan, then I for one would run around MFD opposing deletion of any portal which had a non-spam history. I'd support a moratorium to allow time for a plan to be developed; an RFC like that could enable speedy keeps of any such MFDs.
But the problem is that there is no such of even a seedling.
About 5 weeks ago, I began drafting an RFC. I abandoned that process for several drama-related reasons, but I now think that I was coming at it all from the wrong angle. I thought that the questions I was positing there were fundamental, but in reality it now seems to me that they were secondary. I was looking variants of "which topics should have portals" and "what preconditions should be applied before creation". In hindsight that looks to me like planning a roadtrip without a destination, with no car, and without any agreement on why we are considering a journey.
The big questions now seem to me to be:
What are en.wp portals actually for? Navigation, magazine, or a hybrid reader-editor space? Or some combination thereof?
Do we have models of portal which actually do a good job on any of those criteria? (I don't see any portal which does a good job on any of those fronts. Most portals are atrocious for navigation, poor-to-mediocre as magazines, and poor-to-mediocre as reader-editor hybrids.)
If there is a model of portal which meets whatever standards or goals were set, will readers actually want it? Given the evidence of failure so far to attract readers, we need real evidence that the new model works, not aspirations or ILIKEITs.
If we do find a model which meets all the test above, what do with portals which fail those standards due to lack of curation and maintenance? (For example, should we delete them? Should we have a triaging process whereby the never-any-use are deleted, while portals which fall below the minimum for a period of X months are disabled by being moved to draft space, to avoid luring readers to waste time on them?)
Keep. As a country with hundreds of related articles, this passes WP:POG with flying colors. The question about whether portals should exist at all is a legitimate one, but it is one that should be settled by a new RfC (I believe one is currently in development), rather than an MfD. —pythoncoder (talk | contribs) 20:00, 22 April 2019 (UTC)[reply]
No, this is explicitly notabout whether portals should exist at all. It's about whether there is any way to rescue this particular portal from its long-term pattern of mediocrity: short bursts of improvement followed by long neglect, and consistent lack of interest from readers. It's an attempt to get away from the abstracts, and to ask specific questions: what can the limited pool of editors working on Irish topics do to make this particular portal a) sustainably less rubbish, and actually b) read by readers?
The wider discussion will inevitably focus on principles. But if those discussions of principle are to have any basis in reality, they need to be assessed in conjunction with specific examples. Here's a chance to do some of that. --BrownHairedGirl(talk) • (contribs) 21:52, 22 April 2019 (UTC)[reply]
I see, as I did not add a bolded !vote maybe, I did not count as a Delete. And that's fine - I'm not looking to dump good work, but I do think this debate is necessary. If the supposedly central portal for the Irish family of articles, with 10s of thousands of monthly views, has almost no views, this self-diagnoses as a problem. It is a best a distraction, and it may consume time which an editor would quite happily spend elsewhere. I'm not against the idea of meta-pages, as I already said. Before Wikipedia I spent years volunteering with the Open Directory Project and other "organise the web" initiatives, but I do, truly, believe that such tools have simply passed their moment for most readers.
@Espresso Addict:I like the analogy with publishing, but I think, taken to a logical conclusion, it validates the concern under discussion here. Whatever some people think, making a real magazine that anyone would want to read is hard work. It requires multiple skills, really a whole small team (curation, editing, sub-editing, design and layout, graphics, etc.), and, as one who has been caught with the task for various clubs and societies over the years, I know how easy it is to produce "boring" at worst, "thin" at best. So to offer a good portal is a serious venture. The 2018 initiative to drive out tool-based portals was not a great success, to put it mildly, and it was a grave error to take "don't delete all" to somehow licence "mass creation" - what was need was loving curation of the very best portals, to show what could be done, and inspire others to make working paradigms, if possible. This did not happen, as far as I see, and I've read a lot of portals and their linked articles over the last year. I don't have answers to the general issue, but as for Ireland, I still harbour grave doubts that a reader-attracting and reader-useful portal can be formed anytime soon, given current resourcing and editor interest. And absent that, I still think we might be better without it at all. I leave the debate to others now.SeoR (talk) 08:53, 23 April 2019 (UTC)[reply]
@SeoR: Your opinion should count without needing bolding, but admins are only human, and with an MfD this long & rambling it might get missed.
I recall once writing an entire leaflet overnight on my own in a Portuguese hotel; all the text, all the graphics, all the layout, all the copy editing to a print-ready file... they had good espresso at that hotel... But I think most editors here have computer skills, not publishing/research/graphics/copy editing. And one can't make an exciting magazine-cum-portal that everyone wants to read unless there are articles that support what one wants to say. That really would be a content fork.
I come at all this from the (now retired) featured portal process, where we tried to assess portals for quality. It wasn't a wholly successful enterprise – looking back, I think we were prioritising some of the wrong things – and it seems to have had limited (or no) effect on editors/projects actually keeping the portal up to date, once they got their shiny bronze star.
As I've written elsewhere there's no obvious correlation between quality and readership. No-one's figured out how to drive readers from articles to portals in reasonable numbers. The best suggestion I've seen is to add them to the default search results, but I can't see community consensus developing for that in the current climate. Espresso Addict (talk) 09:29, 23 April 2019 (UTC)[reply]
Another model of portal. Portal:East Frisia is interesting. It avoids the magazine-style "Selected articles" format, and instead has lists of key topics under several headings.
This is a much closer fit to the "gateway" norm of web portals than the magazine format used on so many other Wikipedia portals. It also look like it needs much less maintenance than the magazine format.
But like almost all portals of any format, it gets only risible pageviews: an average og 6 per day in the first 3 months of 2019. --BrownHairedGirl(talk) • (contribs) 01:18, 23 April 2019 (UTC)[reply]
As far as I can tell, all of these German region and state portals were created by a single user and exactly copy the format, subpage structure, and even images (!) of the corresponding portal on German Wikipedia (https://de.wikipedia.org/wiki/Portal:Ostfriesland, in this case)-not sure if that raises GFDL issues. But they all certainly show funky translated grammar and raise structural issues, since these portals all have a subpage like Portal:East Frisia/Members, which is linked at the bottom with the text "The members and supporters of the East Frisia portal are listed here by topic and geographical assignment." Huh? UnitedStatesian (talk) 03:11, 23 April 2019 (UTC)[reply]
You can always refer to the "single user" by pinging me; that would be a kindness. Here are my comments. First, it's true that I've created quite a lot of German regional and state portals (is that wrong?), but by no means all of them. Second, they are rarely "exact" copies of the German portal (but why would it matter?). A couple are totally original and have no German equivalent; most have been tweaked not least to reduce the inevitable red links. And in some cases I've added my own images and articles which are different from those on de.wiki. Of course, there is absolutely nothing wrong with using the German Wiki articles and images if they exist - why on earth not? And if anyone believes there are copyright issues on a bunch of links, it has never been raised before, but is easily resolved by adding the usual link on the talk page. And if you come across any "funky translated grammar" why not just be BOLD and change it yourself? Can I ask if you would accept a portal under any circumstances or is your view that they should all be deleted? Bermicourt (talk) 17:00, 23 April 2019 (UTC)[reply]
Of course, thanks for the question. Portal:Germany and Portal:Alps (and thanks for you excellent work on that latter one) both meet the portal guidelines, in my view, and if they were deleted, whether as a result of a one-off MfD discussion or as a as a result of unfortunate community consensus to reduce the number of portals to a handful or zero, I would be very unhappy. To me, 1,000 portals sounds like the right eventual number; only about half of which (again, my opinion) have even been (re)started yet. So I would expect the MfD discussions of other portals (including potentially of every sub-national political and geographic division other than cities) to continue for quite some time, especially since we are now past mass MfDs and have moved to one- or two-at-a-time. Can I ask you the converse of your question: can you give me examples of topics that you feel are too narrow or insignificant to warrant a portal? UnitedStatesian (talk) 20:03, 23 April 2019 (UTC)[reply]
Thanks, that's helpful. You, me and BHG may not be all that far apart then. If I were pressed, I might say no more than 2,000 eventually, heavily caveated with 'but only if they meet agreed criteria' of which political and geographical levels would be a factor. In fact, you'd probably like the German Wikipedia's rules which are pretty prescriptive and they manage to get all their portal links on one page. I translated their 'relevance criteria' here and proposed an adaptation of them as part of the portal discussions, but they were drowned in the storm surge to create a gadzillion instant portals. My guess would be that, if we could just focus on an initial set of relevance criteria for portals, that would make the culling of low importance portals easier and keep a lid on new ones, especially if we adopt something like de.wiki's approval process. On top of that if we could agree on standards - addressing the sort of questions that BHG raises here - we could also focus on improving existing portals to meet that standard or culling those where no-one is interested (maybe like this one?) and starting again when someone is. To try and answer your question. Firstly, I felt most of the instant portal topics didn't warrant a portal. Secondly, I'd say those that fail the sort of relevance criteria I've just highlighted. If I'm honest, I'd be disappointed to lose the German state and regional portals I've created as I see them as only one level below country level (most of the regions are pretty large and have pan-state, national or even international importance) but if we nailed suitable criteria, I might even reluctantly vote to drop one or two of those lol. Does that help? Bermicourt (talk) 20:42, 23 April 2019 (UTC)[reply]
Hope I can butt in briefly. Portal:Alps seems to me to be v good. A few layout glitches, but it's a meaty navigational gateway. If we are gonna have portals, that's pretty much how I think they should be: lots of grouped links, on a range of topics. Excellent work, @Bermicourt.
Portal:Germany seems to me to be less successful. It's a well-constructed implementation of the magazine format, but like even the best of that breed, it's little use for navigation.
@UnitedStatesian, I agree that most of the portals which would be worth keeping have probably not yet been created. But I think that the viability of any given number depends largely on whether the format used has maintenance requirements which don't exceed the capacity of interested and appropriately-knowledgeable editors. E.g. Portal:Ireland is not viable in its current form, but might be viable in another form. So if the chosen format is old-style wholly-manual magazines with forked excerpts, then we might not be able to sustain 100 in decent condition; but other formats require vastly less skill and effort to maintain, so we might hit 1,000 of them. But even the excellent Portal:Alps gets a median of only 6 pageviews/day. --BrownHairedGirl(talk) • (contribs) 21:17, 23 April 2019 (UTC)[reply]
@Bermicourt: That link to your translation of the German criteria is very helpful; thanks so much for doing that and sorry I missed that earlier; I expect it will get a lot more attention as we continue our discussions. My one comment is, that would lead to a lot more than 2,000 portals. Just on geography alone, you've got ~200 countries, say each one has a capital and an avg. of 18 next-level-down political divisions: that's 200 x 20, or 4,000 right there. The city criterion (I'm guessing) adds 4,000 more. With that 8,000 just to start you get far north of 10,000 very quickly. I just don't think en Wikipedia will ever be ready for that. Maybe come at it the other way: ask this community, "How many portals is the right number?" and then back into the criteria that get us to that number? UnitedStatesian (talk) 13:30, 26 April 2019 (UTC)[reply]
@UnitedStatesian: Yes, you're right we need to do the maths as well and use that to adjust the criteria. Also there are considerations like - should every country be treated equally? Should a tiny city state get the same coverage as Russia, etc? We keep quoting reader hits and 'broadness' which are crude gauges of topic notability; maybe they need to be refined. So I see the German criteria simply as an example of how we might tighten up the rules here. Bermicourt (talk) 14:32, 26 April 2019 (UTC)[reply]
Keep but improve, taking account of BHG's valid points, but only once the community has agreed standards. The scope of this portal is easily wide enough as others have said. But BHG's criticisms are worth taking on board, not just here, but across the portal spectrum as part of the much-needed review of purpose and standards. At the moment we're doing this in a vacuum. What is improvement? And when will it be good enough? We don't really know because there are currently no agreed standards for portals (WP:POG is vague and out of date), so editors are just going with their gut feelings, hence we're swinging from one (TTH) extreme - create thousands of pointless portals - to another extreme - delete most if not all portals. And the end result depends on how many pro- and anti-portaleers can be mustered to outvote the opposition. It's not a great way to work. Instead we should stop the infighting until we can agree what the purpose of portals is and what standard they should attain. Bermicourt (talk) 17:00, 23 April 2019 (UTC)[reply]
@Bermicourt. I agree that there is a vacuum. Not just no agreed standards, but no agreed goal which standards might assess.
The problem here is that ENDPORTALs produced a classic variant of incumbency effect. It's extremely unlikely that a proposal to create from scratch a collection of portals would have passed without any clear understanding of what it is actually for ... but the deletion proposal united in opposition those with any vision of what a portal might be. Magazine? Nav hub? Picture gallery? Chance to play with fancy coding? Space to immortalise that DYK you made in 2013? Billboard to advertise your WikiProject? Nostalgia for Yahoo c. 1997? One of en.wp's few opportunities for full-page web design? Keep the portals namespace and your goal is still in play, even if nobody shares it.
The only roadmap on offer was the WP:TNT, but that was rejected. So ENDPORTALS closed with no roadmap, and the opportunity to create one was basically grabbed by the portalspammer, who drove the whole Portals Project at high speed into a cliff-face. Splat.
ENDPORTALS agreed not delete everything, but it also precluded nether mass expansion nor deep culling. So now we have that vacuum. And nature abhors a vacuum.
Unless and until that vacuum is filled, it's perfectly reasonable for editors to look at a portal page and decide that it has little value, no direction, no minders, and therefore no future ... so they will MFD it to weed it out just as a gardener will do with a neglected garden.
The plea by you and others who see value in portals is hey "hang on till there's a plan". In principle that's reasonable enough, but given the history it's also reasonable to respond that the whole problem is that there never has been any plan apart from last years madcap automate-and-expand-and-forget-the-quality scheme ... so why wait for one when the odds look so slim?
(It's a bit like the problem faced by Fianna Fáil in the 2011 election, when their pitch had to be something like "sure, we were corrupt as hell, so incompetent that we didn't even have a PhD-level economist in the Dept of Finance, spent too much time partying in that Galway tent, turned our think-ins into drink-ins, overlooked key economic indicators that were handily available even on Yahoo.Finance, and drove the country over a cliff for the 3rd time in 80 years ... but sure, vote for us like your grandparents did, and we'll think of something". It wasn't an easy sell, and it didn't work. they bounced back a bit a few years later, but it takes time to regain a passable level of credibility after a fiasco.)
So there's no chance of a moratorium being agreed unless there's at least a new process towards a new consensus. There are some people associated with the portals project who I'm v happy to work with, such as @Bermicourt and @Espresso Addict; but there are several others who think would give me migraines. That's partly why I tried launching a discussion here rather than going directly towards RFC design as Bermicourt and I had discussed elsewhere.
What's the next steps? Should I draft an RFC along the lines of the 4 points I listed above in response to Espresso Addict? --BrownHairedGirl(talk) • (contribs) 18:58, 23 April 2019 (UTC)[reply]
That sounds like a good idea, albeit I confess I'm struggling to identify which 4 points you're referring to. I also like Espresso Addict's idea that we develop a 'model' portal to showcase what we mean. Finally, I just want to say that much of the debate seems to focus on portals as a reader tool and the (low) number of hits they get. While I think that needs addressing, the use of portals as a tool for editors to improve and extend article coverage gets forgotten. So we need to remember both. Bermicourt (talk) 20:42, 23 April 2019 (UTC)[reply]
Sorry, @Bermicourt, look for the timestamp "16:51, 23 April 2019". My 4 points are immediately above that, in a reply to Espresso Addict.
[ec] This is a good question. It gets to the core of the problem with portals, which is that it is not clear what they are or could be useful for, or what constitutes a useful portal. Or even whether these things should or can usefully be dictated by a set of rules. The hit rate suggests that they are not particularly useful to the average reader, which may be because the average reader does not know about them, as they are not particularly well linked from mainspace. I don't know whether they are returned in searches, but I suspect not. As BHG stated above the search engine has largely superseded navigation aids like satnav has largely superseded road maps, but search engines find what you ask them to find, not necessarily what you are looking for, and you may not know how to describe what you are looking for. A search engine is great when you know the right search string, but can be amazingly unhelpful when you are looking for something related but are not sure what it is called. A good navigational aid can provide a topic structure in a way that helps the user find things they know should be there, but don't know what they are called, and things they didn't even suspect may be related. This is not an easy thing to create or display in user-friendly format, but it is what I would like portals to be. It is not necessary for them to be called portals, or be in portal space, or even to be permanent code - it is the functionality that I think would be useful. Eventually this functionality may come to search engines, but until then I see value in experimenting with methods of presenting related topics that show the inter-relatedness of content relating to the topic of interest, and the extent of available information on Wikipedia connected with that topic in less obvious ways. I dont think that any of the current portal formats are much good at this function, but there is a chance that by experimenting with them we may get something that does the job. Restrictive rules on how portals must be structured may block potentially useful lines of enquiry. High maintenance structures are unlikely to develop into anything more widely applicable, and are more likely to fizzle out when interest wanes. Whether this is actually a problem to anyone except the creator and maintainers of the specific portals is a separate matter. My opinion is that it is a trivial matter, as the time expended is what the user chooses to expend. It may do little good, but also does little harm. Rules for portals should be based on what we want portals to become, and should not unnecessarily restrict how they get there. On the other hand, a minimum standard for functionality/utility for a portal that is not actively under development is advisable. I am sure that there are many who have different visions for portals, and I don't see any fundamental reason why many of them cannot be accommodated in a reasonable set of criteria. I would personally like to see as many different approaches taken as people are willing to experiment with. Most will probably be dead ends, some may have to be culled as counterproductive, but some of them may evolve into something better. That's life.
With regard to the example of Portal:Ireland. I think it is easily a broad enough topic to have a portal, but that the current portal does not serve much useful purpose. It also does no perceptible harm, except possibly as a mild embarrassment, and the content may be useful as a basis for something better. I would not !vote to delete, but would also not oppose deletion. · · · Peter Southwood(talk): 06:49, 24 April 2019 (UTC)[reply]
BrownHairedGirl Your four points look like a good place to start. They may be possible to fine tune a bit, but they address the fundamental problems fairly well. If we can get actionable answers to those questions we will be far ahead of where we are now. · · · Peter Southwood(talk): 07:00, 24 April 2019 (UTC)[reply]
BrownHairedGirl In your second point you state that most portals are atrocious for navigation. In general I agree with you, but I did try to make a portal that was specifically better for navigation. I am aware of your opinions on portals based on a single navbox, but would like your opinion on whether the combination of Portal:Underwater diving and its collection of single navbox subportals was any better as a navigation aid than others. Whether it or they survive deletion is of secondary importance, and I would prefer to keep that aspect out of this discussion. I think that navboxes have a potential for navigation that has not been fully developed, and there may be some way to use combinations of closely associated navboxes to guide readers to articles related to the broader topics in a way that goes beyond regular wikilinks, which only cover topics mentioned in a specific article, and search engines which rely on the right choice of search string, and single navboxes which are limited in size to avoid overloading the article, or even portals as they are often constructed with more of an intention to showcase quality than illuminate the full scope of a topic. At present I have my doubts about category trees, the only other available system that is sufficiently inclusive and versatile, because they tend to gather low relevance clutter such as maintenance categories, which are useful to editors by not so much to readers. I have the seed of an idea for using a nested set of normally collapsed navboxes which could rely on mouseover for summary content display, but don't know how well it would scale to large topics with potentially thousands of articles. This is rather tangential to portals as such, but may have some relevance to what we want portals to be. Anyway, I plan to think about it a bit, maybe build a prototype and see how it works. Cheers, · · · Peter Southwood(talk): 20:20, 26 April 2019 (UTC)[reply]
Keep for now. Kinda pro-forma, but since this page was intended only as a trigger for discussion and not an actual delete-now proposal, I should probably have added this at the outset. Sorry.
I am personally delighted with the great range of very thoughtful contributions from a variety of editors who come from very different angles. For me it has been the most productive discussion so far on portals, and already seems to have crystalised a few ideas for wider meta-discussions, which I hope will lead to a new RFC. Many thanks to all participants. --BrownHairedGirl(talk) • (contribs) 17:01, 24 April 2019 (UTC)[reply]
Keep this is wild; deletionists have gone too far. You can't just delete a portal because 'not good enough'. Try finding an actually valid reason. ɱ(talk) 17:13, 24 April 2019 (UTC)[reply]
Redirect to Ireland, a method of deprecating and archiving. No case for deletion, this discussion does not belong at MfDeletion. Redirect to Ireland because it forks from the article, detracts from not supports the article, is massively more an editor cost than a resource enjoyed by any reader. It is worse than forked content, as it ignores the explicit sourcing requirements for encloepdic content, and reader who lands on it on arrival may adopt 2004 thinking on how to build content. At a navigation aid, it fails, failure demonstrated by pageviews, and its navigation mechanisms redundant to the article’s wikilinking, categories, and the (over)abundance of navigational templates. It is a lot of work for an unwhelming result for editors, and a negative impact for readers. Typical of virtually all portals, if not all, its day passed over ten years ago and all should be archived. Redirection of each to its parent article is my recommended method of archiving. —SmokeyJoe (talk) 02:18, 25 April 2019 (UTC)[reply]
I would prefer marking as historical with a redirect to Ireland to a hard redirect as a form of archiving if that were to be the decision. · · · Peter Southwood(talk): 20:20, 26 April 2019 (UTC)[reply]
I agree. Archive & soft redirect to the article. —SmokeyJoe (talk) 21:28, 26 April 2019 (UTC)[reply]
Comment. I know this is straying off topic a bit, but all the interesting current discussion seems to be happening here... I've just been looking at how German Wikipedia uses portals and noticed that e.g. Portal:Bahn gets twice as many views as Portal:Railway even though German Wiki presumably has a smaller audience. On checking how they're linked to mainspace, I noticed that Portal:Bahn has relatively few article links (only the main topics plus a few random articles where editors haven't followed the conventions), but appears to be linked from all the category pages. In other words, it is being used in a different way - more as a tool to complement (and sit above) the categories (as well as being used to aid WikiProjects). I've only had a quick look, so more research required, but that may be a better way to go as far as the reader-facing usage is concerned...Bermicourt (talk) 14:38, 26 April 2019 (UTC)[reply]
Keep. I personally do not particularly like portals , and I have never helped work on any. This goes for a variety of other organizational devices here, but we're not making WP to suit my own immediate interests. So unless they are to be eliminated totally, and there has clearly not been consensus for that, the most useful or potentially useful, ones should be kept. This has the potential of being among them. It it proves otherwise, we can revisit in a year or two. DGG ( talk ) 09:07, 27 April 2019 (UTC)[reply]
The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the page's talk page or in a deletion review). No further edits should be made to this page.