You don't need technical skills in order to contribute here. Be bold when contributing and assume good faith when interacting with others. This is a wiki.
This is a deletion request for the community to discuss whether the nominated page should be kept or deleted. Please voice your opinion in the linked request above. Thank you very much!
If you created this file, please note that the fact that it has been proposed for deletion does not necessarily mean that we do not value your kind contribution. It simply means that one person believes that there is some specific problem with it, such as a copyright issue. Please see Commons:But it's my own work! for a guide on how to address these issues.
Please remember to respond to and – if appropriate – contradict the arguments supporting deletion. Arguments which focus on the nominator will not affect the result of the nomination. Thank you!
I'll mark all 7 old images with "speedydelete" and the same reason they now have each one of them. I was the original uploader of all 7, and have changed all uses in Commons and Wikipedia to the new JPEG equivalents. I would have liked to upload a new version for all them, but it is not possible to replace an image with another in a different format. I could not create (original files were in PDF format) new PNG images with good quality and reasonable size, so I had to make JPEG ones. MGeog2022 (talk) 13:48, 13 September 2022 (UTC)Reply
@Jeff G.: They were of lower quality themselves, as compared with the new JPEG ones (it's easy to see it when viewed with enough zoom). I made a mistake using PNG for this kind of image at first. I'll add "F8. Exact or scaled-down duplicate" to all of them. MGeog2022 (talk) 14:08, 13 September 2022 (UTC)Reply
This is a deletion request for the community to discuss whether the nominated page should be kept or deleted. Please voice your opinion in the linked request above. Thank you very much!
If you created this file, please note that the fact that it has been proposed for deletion does not necessarily mean that we do not value your kind contribution. It simply means that one person believes that there is some specific problem with it, such as a copyright issue. Please see Commons:But it's my own work! for a guide on how to address these issues.
Please remember to respond to and – if appropriate – contradict the arguments supporting deletion. Arguments which focus on the nominator will not affect the result of the nomination. Thank you!
Latest comment: 8 months ago7 comments2 people in discussion
Hi!
As many data on the web vanished (partly due to political motivations), I would like to ask you if you know useful datasets that could be uploaded to Commons? I am covering orthophotos of Germany right now, but I can imagine you have some good ideas, too :)
Hello @PantheraLeo1359531. I don't know many details about the US datasets that are currently in big danger, but I suspect that they are mostly numeric or textual data that I don't know how they would exactly fit into Commons (but they probably could). It's great to have freely licensed orthophotos and maps about one's own country. It's also the case with, for example, Spain and the United States, but, unfortunately, not with countries such as the United Kingdom.
There are some cases where I dream about having full collections in maps uploaded to Commons, but I think about it as something almost impossible to do it by myself, due to the needed time and effort:
Spain's National Topographic Map (source: https://centrodedescargas.cnig.es/): I've uploaded the maps for the main cities and several interesting areas, for the first digital edition of MTN50 (1:50,000 scale). It would be great to have the more than 1,000 sheets that form that edition of the map (a unique edition because it shows full Spain consistently as it was in the 1999-2010 period, while other editions, both pre-digital and the next digital editions, are updated only on a per-sheet basis, in very different dates). In addition, it would be even greater to have all the sheets of all editions of MTN50, and the same for MTN25 (1:25,000 scale). There are lots of sheets of MTN25 in Commons, but probably not all from all editions.
Except for the last case, fully uploading (including setting the descriptions, etc) any of these sets of maps is a really formidable work, but using automated tools perhaps could make it possible. MGeog2022 (talk) 12:40, 27 March 2025 (UTC)Reply
Thanks! When I have time and resources, I may upload some of them :)
Yayy :) Digital preservation is a very important topic to me, and I am horrified how people do not care about this topic, let it be in my hometown and elsewhere :(. If you have new ideas, please let me know :) --PantheraLeo1359531 😺 (talk) 19:55, 27 March 2025 (UTC)Reply
@PantheraLeo1359531, an update on this: I uploaded the remaining sheets from the 2010 orthophotomap of Asturias in PDF format, and it's fully uploaded by now. More sheets from the first digital edition of MTN50 (1:50,000 scale) have been uploaded by me and other users, but many others still remain without being uploaded. Regarding other countries with freely licensed maps, I'm sure many is still to be done, but, for Switzerland's case, it's already done, and the result (I have not contributed to it at all) is really great. MGeog2022 (talk) 11:02, 28 September 2025 (UTC)Reply
I think it needs the effort of many OpenRefine-fluent users to upload the humongous amounts of files, as we face several 100,000 files and many many Terabytes in total. Thank you for your efforts :) --PantheraLeo1359531 😺 (talk) 11:45, 28 September 2025 (UTC)Reply
Latest comment: 4 months ago41 comments3 people in discussion
Hi!
I stumbled across Mapillary some weeks ago. It is a free panoramic viewer of street imagery. As it covers also streets of many countries, even in special areas, I would like to know what your opinion is on backing up this content. As Mapillary is owned by Meta, it could be subject to restrictions or more in the near future. It covers many many million images, many also useful for Commons. Kind regards, --PantheraLeo1359531 😺 (talk) 08:20, 28 September 2025 (UTC)Reply
Hi, @PantheraLeo1359531! I didn't know about Mapillary: I wished that something like "OpenStreetView" existed, and it seems that it already existed :-D
Of course, if all or part of its content is freely licensed, my opinion is that it would be great to upload it (or the freely licensed parts) to Commons. If it's several TB in size, it would be good to talk first about it on Village Pump or the like, though. My opinion (and I hope the opinion of most Wikimedians also) is that all free educational (including historical, not necessarily old) media has its place in Commons, unless it's too much burden for Commons total storage size (we don't know what's "too much" without asking: maybe doubling current Commons size is a problem, maybe it isn't).
In this case, a solution would be to prioritize the images considered as most important. Another point would be to categorize the images so they can still be "navigated" in some way, in Commons. Depending on how many people are interested, it would even be possible to create a navigation viewer plugin in Commons (similar to the current ZoomViewer), using OpenStreetMap and the uploaded Mapillary images. MGeog2022 (talk) 10:51, 28 September 2025 (UTC)Reply
I had a talk in a Telegram chat of Wikimedians. They were also in favor of uploading content, but the problem is that some sequences may have unnecessary images (unsharp, wrong focus, etc.). I assume that the WMF can handle some petabytes without problems. I think it would be good to have a crawler that crawls the sequence IDs (in a certain area) to download the high-res (through the API) files and to remove the low-quality ones before upload. Especially high-populated places have incredibly much images. Even my hometown (<50000 inhabitants) has maybe 15000 to 20000 files :) --PantheraLeo1359531 😺 (talk) 11:49, 28 September 2025 (UTC)Reply
@PantheraLeo1359531, good luck with it, it sounds as a really formidable work. Yes, WMF, with its budget, could host several PB for sure (including backups and redundancy), but it has to first take the step to prioritize investments in expanding the storage capacity.
Mapillary images eventually making more than half of total Commons size sounds a bit strange, but, given than many movies and other filmings will enter public domain over the next years, Commons total size will grow a lot for sure, so it's unlikely Mapillary will take more than half of the space for much time. I hope WMF commits strongly to Commons needs: media files take a lot of space, but they are much needed, both to complement the content of Wikipedia and other wikis, and for preservation of historical materials. MGeog2022 (talk) 18:22, 28 September 2025 (UTC)Reply
@PantheraLeo1359531, for several million images, yes, it makes sense to think about automatic batch uploading :D
While being data from a single source, that could make more than half of total Commons size, I think it's good to have it here despite that, because it's like a "view of (parts of) the world", that wouldn't be otherwise possible to have available in Wikimedia. That is, it's a really big amount of data, but also a very important one.
By the way, whether it's eventually added to Commons or not, it would be fine to also include it in OpenStreetMap as a kind of "OpenStreetView", if OpenStreetMap is open to that (and it would help to expand the coverage of the street views, with the help of OpenStreetMap community). MGeog2022 (talk) 12:52, 1 October 2025 (UTC)Reply
Fun fact: Afaik one similar project was named OpenStreetView but requested to change its name :D. But you're totally right. We need to connect the data between Mapillary, OSM and Wikimedia Commons, preferably with redundancies to avoid data loss. The advantages are clear: Digital twins as public good (accessible for everyone)[1], records of past structures, educational value and structured data, can be combined with OpenGeoData by (for example) German "Landesämter" (like lidar measurements, orthophotos, maps, etc.). And we need a strong free alternative to all the commercial services who set up a huge anti-Commons (unfortunately with the help of many million people who gift their imagery to them for free). Being dependent on commercial services is like sitting on a twig that is about to burst, but you don't know when exactly. ArchiveTeam claims that Mapillary has a few petabytes of footage. Of course it's a lot, but not impossible. The big challenge will be in the near future to change the opinion of the society towards free and open data while companies shamelessly accessing more and more (personal) data (you certainly know about the fatal actions of US government and companies like Palantir, digital feudalism, etc.). When I tell people what I do, many don't see the importance or may look irritated at me😕.
[1] At one fair, we had representants of the district of my hometown and they showed the advantages of digital twins (for example putting in a possible new building into the environment that could be built there, etc.). Many people don't know about the advantages and then claim there are none. --PantheraLeo1359531 😺 (talk) 13:48, 1 October 2025 (UTC)Reply
@PantheraLeo1359531, yes, that's clear: commercial companies have no obligation to preserve the data they store, and they only store it while it makes money. That's the advantage of non-profit foundations such as WMF or Internet Archive (in the latter case, if they properly backuped their data, which I doubt, as I've expressed in Village Pump), because they are commited to keeping the data in perpetuity.
When I tell people what I do, many don't see the importance or may look irritated at me: you can just tell them to imagine what would happen if that data vanished tomorrow. For example, Google has the legal right to close Street View at any moment... yes, it's very profitable, but maybe the historic images aren't, so they could delete them to save space. "Good vs bad" rethoric may not persuade some people, but it's just about purpose: commercial companies are created to make money, while certain non-profits are created to preserve knowledge, and both have their place (some companies create freely licensed content that can be preserved by the non-profits, in the same way many companies create free software that can be preserved by, for example, the non-profit Debian project).
the advantages of digital twins: this includes preserving how streets or buildings looked at a certain time. With notable buildings, this is usually guaranteed. If you like your own (past or present) house or street, you should like that its present look is as preserved as the ones from notable buildings are. MGeog2022 (talk) 13:23, 2 October 2025 (UTC)Reply
There are good points :). Yeah of course, how many people like to see what their neighborhood looked years or decades ago :D. I think the trend on generating automated imagery of streets etc. will increase, but the question of preserving them will be huge issue. Many, many exabytes will be created and need to be stored, I really hope that concepts by public libraries and archives will be set up soon to combat this issue. Basically, it's often OUR data for data FOR US, so it needs to be treated like that :) --PantheraLeo1359531 😺 (talk) 18:19, 2 October 2025 (UTC)Reply
@PantheraLeo1359531, thanks for the info! It's great to know that street images from around the world will be available in Commons thanks to your work (and, probably, also the work of others).
In your first message here about this topic, you talked about imagery of Greenland. Yeah, that's a good starting point just now. Just have a look here... I'm sure it's the one getting the most visits this week, out of all the files uploaded by me. Greenland is a guarantee of success at this very moment :-D MGeog2022 (talk) 13:50, 20 January 2026 (UTC)Reply
@PantheraLeo1359531, really wonderful, wow! Thanks. The only problem is that the images are not categorized, other than the hidden categories (I thought categorization would also be automatic). Unfortunately, I've added 722 images to the list of uncategorized files currently in Commons :-( MGeog2022 (talk) 20:37, 20 January 2026 (UTC)Reply
Happens ;). Fixed it! I use Mapillary ({{mapillary.user.username}}) {{captured.date}} {{captured.hours}}H{{captured.minutes}}M{{captured.seconds}}S{{captured.milliseconds}} ({{mapillary.photo.id}} at {{mapillary.photo.sequence}} with {{camera.make}} {{camera.model}}).jpg as name template (to include any information in the title name, and usually the categories: [[Category:Photographs of <CITY> from Mapillary (<SEQUENCE ID>)]] ::::::::::[[Category:Files from Mapillary uploaded by PantheraLeo1359531]] ::::::::::[[Category:Images from Mapillary uploaded with Curator]] ::::::::::[[Category:Files uploaded by PantheraLeo1359531]], while there are my usercats. Maybe this helps :) --PantheraLeo1359531 😺 (talk) 21:22, 20 January 2026 (UTC)Reply
@PantheraLeo1359531, many thanks! Could you give me a step-by-step about where "Category:Photographs of <CITY> from Mapillary (<SEQUENCE ID>)" is to be placed, in the tool? Thanks in advance. It's the only part I need before doing more Mapillary uploads (since "Category:Images from Mapillary uploaded with Curator" is automatically added, and I don't use nominal categories for my uploads). This is a very interesting tool :-) MGeog2022 (talk) 22:33, 20 January 2026 (UTC)Reply
At point 3 EDIT, there is a section called "Fallbacks". Inside it, there is a textbox with a grey "Categories". This is basically the area where you can add wikitext (like [[Category:Photographs from Mapillary etc.]] or FoP-Templates). The "Exclude from date category" is useful if you manually add [[Category:<COUNTRY> photographs taken on YYYY-MM-DD]]. Then it does get categorized in the country photographs category, not the mothercat :). Greetings --PantheraLeo1359531 😺 (talk) 10:23, 21 January 2026 (UTC)Reply
@PantheraLeo1359531: I've had a problem: I missed the "Category:" part in an upload. Is it possible to add "Category:Photographs of New York City from Mapillary (e6dm23f8qovg9our24i9z7)" to all the photos I'm uploading now, replacing the "Photographs of New York City from Mapillary (e6dm23f8qovg9our24i9z7)" text? I'm not familiar with Commons automation tools. That's the problem with mass uploads: errors are not manually fixable. I'll try to be more careful in the future. Thanks in advance (please wait until about 1 hour from now, at least, so the upload is complete). MGeog2022 (talk) 15:10, 21 January 2026 (UTC)Reply
@PantheraLeo1359531, so is it possible to fix it, isn't it? Thanks for fixing it when the moment arrives, and thanks for the info about the button (I won't cancel now, since I understood that it's possible to fix the category issue). I also suppose that if I log out the mass upload that I started will continue until the end, isn't it? MGeog2022 (talk) 15:29, 21 January 2026 (UTC)Reply
Confirmed: the upload goes on after logging out (this makes the tool even greater!). So time to take a rest, and, when using it again, I will be more careful. Now that I know that I don't need to be looking at the upload as it progresses (and that, probably, I can start several parallel uploads without problem), I'll dedicate more time to looking twice at the categorization :-) MGeog2022 (talk) 15:37, 21 January 2026 (UTC)Reply
Done Don't worry, mistakes happen ;). I fixed the categorization with VisualFileChange. I filed a request to automatically add the milliseconds to the title, because sometimes 2 or more images are taken within one second, and they are uploaded to the same file name. The millisecond timestamp makes every filename unique. Greetings --PantheraLeo1359531 😺 (talk) 19:35, 21 January 2026 (UTC)Reply
Well, probably, I alarmed myself too much :). You clarified it well enough: for panoramics, de minimis should work here. But it's important that such Mapillary photos are panoramic (many others aren't, and I wasn't thinking about FoP, so many thanks for the warning: uploading 700 copyvios is not funny).MGeog2022 (talk) 14:41, 23 January 2026 (UTC)Reply
Yeah, the problem is really annoying, so I wanted to warn soon enough :D. I thought about panoramics from Syria, but the similar problem here. For panoramics (especially yours from France), I don't see any problems, but other pictures from Mapillary may be. I focus on pictures from FoP countries for architecture, which are quite a lot :). [[COM:FOP <COUNTRY>]] usually gives advice on what from which country is okay to be depicted :). Greetings --PantheraLeo1359531 😺 (talk) 15:19, 23 January 2026 (UTC)Reply
Hi, thank you for making this tool! But can you please upload it somewhere where its easier to find, since i had to look at this user talk page to find a very helpful tool which should (in my opinion) be somewhere for everyone to find. Thanks ! CubanoBoi (talk) 01:01, 26 January 2026 (UTC)Reply
@CubanoBoiHi! Yes, the tool is in a sort of Beta stadium, and there are still some things to figure out, especially when it comes to a mass usage by many users at the same time. This is why I wanted to perform a smaller rollout ;). But I will talk about how to make the next steps. --PantheraLeo1359531 😺 (talk) 09:52, 26 January 2026 (UTC)Reply
@PantheraLeo1359531, thanks for the info! And, the most important news, that data now seems to be really preserved. WMF has proper media backups since around 2021 (before that moment, it hadn't, and, sometimes, bugs resulted in data loss), and Internet Archive has backups "around the world", according to a 2025 CNN article. I don't know when they did start, but data from 2016 showed that they had only 2 copies, both in San Francisco area; that's why I was so worried about it and I talked about it several times in Village Pumps (both Commons and Wikipedia), before I knew about the good news.
As for those new 50 TB, without my Mapillary mass uploads this weekend (I had to take advantage of my free time on the weekend), they would probable be 49 :). It was a hard work, but we now have photos from many countries, uploaded mainly by you and me, as I could see in the tool's history. I don't know if the tool is still in a stage of "beta testing", but maybe it would be good to talk about it in Village Pump, so many other users can join us (perhaps restricted to autoconfirmed or autopatrolled users, or some other measure to prevent miscategorization by unexperienced users). Both the tool and our many uploaded files deserve to be better known! MGeog2022 (talk) 20:19, 25 January 2026 (UTC)Reply
Indeed! The tool was coded by DaxServer and became public in November 2025. I was allowed to test it and make suggestions. I think it is already quite mature, but I am careful to make it very public, because it could put the tool under pressure when it has too much users :D. Thank you for covering major cities! Maybe we can find a proper solutions in following discussions.
@PantheraLeo1359531, I think a potential problem is that a non-autopatrolled user can quickly gain the right to become autopatrolled (>500 edits) by using the tool. But, probably, it's not the only way to achieve this by using mass uploads. I hope admins also look at the quality of the edits and not only the numbers :-) MGeog2022 (talk) 13:48, 26 January 2026 (UTC)Reply
@PantheraLeo1359531, it looks very good. Only a small suggestion to improve it: include also how to get the sequence key, step by step (clicking the circle with the three dots, etc). I've just made another improvement: providing a direct link to the tool, so my talk page is no longer the primary way to get it :-D
Another suggestion: mention that the batch upload continues working even if the user logs out. At the same time, ask the user to be patient, since there is a single common queue for all the users. MGeog2022 (talk) 20:41, 31 January 2026 (UTC)Reply
I added a new little feature: You can now categorize Mapillary images by date. If you uploaded a batch, you can add the Category Mapillary photographs taken on YYYY-MM-DD to the batch category. There you enter {{Mapillary photographs taken on navbox}} :3. All the best! --PantheraLeo1359531 😺 (talk) 19:57, 3 February 2026 (UTC)Reply
@PantheraLeo1359531, I see you added date categories to some of my Mapillary uploads, thank you. Sorry if I am too lazy, but creating those categories every time I do a Mapillary upload, seems too much work to me (many times, one needs to create the country, administrative division, city and sequence categories, and to link to the "photographies of ..." or main category for several of these items). Maybe I will dedicate some day to create and assign lots of date categories to the sequence categories created by me, but such a day hasn't arrived yet :-D MGeog2022 (talk) 14:26, 8 February 2026 (UTC)Reply
I just thought it is funny to have chronological sorting and how many dates can be collected :). The oldest image I found on Mapillary is from 2002. Some image are tagged with 1989 or even 1970, but this is not correct or valid --PantheraLeo1359531 😺 (talk) 16:04, 8 February 2026 (UTC)Reply
@PantheraLeo1359531, yes, no doubt it is very useful to have them organized by date when possible. I don't know if older photos are allowed in Mapillary (if that's the case, even those 1970 photos could be correct). MGeog2022 (talk) 16:07, 8 February 2026 (UTC)Reply
They seem to be allowed, but the 1970 photos were made by digital cameras on Mapillary, but there were no (high-res) digital cameras ;). Afaik they were claimed to be taken on 1970-01-01 --PantheraLeo1359531 😺 (talk) 16:34, 8 February 2026 (UTC)Reply
Coorect :). I think the number will grow even faster, soon. The developer also examines some options to speed up procedures, this would be necessary. Imagine having 10+ users acting at the same time... 🫠 --PantheraLeo1359531 😺 (talk) 13:40, 12 February 2026 (UTC)Reply
Latest comment: 3 months ago5 comments3 people in discussion
★Your nomination has been reviewed and promoted★ Congratulations! The image you nominated was reviewed and has now been promoted as a valued image. It is considered to be the most valued image on Commons within the scope: 3D model of a fused grid district. If you would like to nominate another image, please do so at Valued images candidates.
Latest comment: 3 months ago24 comments2 people in discussion
Hi! I remembered your suggestions for archiving. USGS offers an S3 url with many files. (https://prd-tnm.s3.amazonaws.com/index.html). I don't know if all files are public domain. I assume we have to check this via the Sciencebase, like here, when we look up the originator. Maybe I find a way, then I could mass import the files from there. I can imagine, you have some ideas on that topic :) --PantheraLeo1359531 😺 (talk) 18:06, 13 February 2026 (UTC)Reply
Hello, @PantheraLeo1359531! All USGS files should be in the public domain, since they are works by the federal government of the United States (see also the article on USGS itself). I don't know if there could be a rare case (for example, if the making of a USGS map was contracted to a commercial company, or the like) of a non-public domain USGS map, but, if exists, it should be a very rare exception.
The total number of sheets, multiplied by all the editions, must be really formidable, but there are already a good number of them in Commons (root category), and many of them seem to be quite recent uploads. Great if they can be mass imported like with Mapillary :-) MGeog2022 (talk) 22:38, 13 February 2026 (UTC)Reply
Nevermind, I just saw that the "The National Map" imagery is indeed all public domain as in The National Map. Okay, I will file a whitelist request :). I just have to think about a proper categorization. Maybe by US state or something... --PantheraLeo1359531 😺 (talk) 09:44, 14 February 2026 (UTC)Reply
@PantheraLeo1359531, in addition to The National Map (mainly edited by volunteers, I see), there are many map sheets downloadable from
https://www.usgs.gov/, covering many editions for each sheet. Sadly, the quality of the USGS topographic maps degraded around 1992 (according to the Wikipedia article about USGS, the best standards were those used between 1945 and 1992). Just compare this sheet from 1966 and the 2019 edition of the same sheet. But, with better or worse quality, having the full historical series in Commons would be very useful to watch the historical evolution of the places.
Nearly 57,000 individual maps in this series, multiplied by several editions, would result in a really huge number of files to upload to Commons, but, for example, just now we have 261,906 files Mapillary images uploaded with Curator, so it wouldn't be impossible at all ;-) MGeog2022 (talk) 18:26, 14 February 2026 (UTC)Reply
57,000 files doesn't seem to be a problem :). I can put more than 100k files in one table for OpenRefine, so this seems doable :D. Do you even have a specific USGS URL to the direct downloads? --PantheraLeo1359531 😺 (talk) 18:30, 14 February 2026 (UTC)Reply
@PantheraLeo1359531, sorry, I don't have any specific URL, I believe it's years since I last downloaded a map from there :-(
57,000 is the number of sheets, but, multiplied by the total number of editions (I suppose it varies from sheet to sheet, at least, that's the case with Spain's national topographic map), it will be a far higher number.
About Mapillary: in countries without freedom of panorama, I only find panoramic images. Maybe it's just coincidence, but perhaps one doesn't need to worry too much about FoP status of each country, so (possibly Mapillary itself is already taking care of it). MGeog2022 (talk) 18:37, 14 February 2026 (UTC)Reply
Alright! I will look through it, when I have some capacities :).
@PantheraLeo1359531, it makes sense that they care about it: since all Mapillary images are licensed under CC-BY, in non-FoP countries, such licensing would violate copyright when a recent building is shown and de minimis isn't applicable. MGeog2022 (talk) 18:44, 14 February 2026 (UTC)Reply
I've just found some non-panoramic Mapillary photos from Reykjavik (Iceland is a non-FoP country), so my theory can be discarded. I don't know the legal basis for such CC-BY licensing: maybe all buildings are shown in a de minimis way even while the photo is not a panoramic, but I won't take any risk. MGeog2022 (talk) 18:53, 14 February 2026 (UTC)Reply
The question is, what is considered here. Mapillary can say the license only covers the image itself, not the depicted motifs. Maybe the awareness of FoP is not there. I wonder how this issue is managed by Google in Google Street View --PantheraLeo1359531 😺 (talk) 19:28, 14 February 2026 (UTC)Reply
PS: I had a talk with a WMDE member in the last days, and she said that WMF wants to focus more on Commons. This is really needed, because we still have some important actions and projects to take, as the textured 3D models, for example --PantheraLeo1359531 😺 (talk) 08:55, 15 February 2026 (UTC)Reply
she said that WMF wants to focus more on Commons: that's really good news. Commons is the WMF project that needs the most storage resources (for obvious reasons: the other projects only store plain text, with some exceptions), so it's good that WMF values its importance. Also, Commons, in addition to being a needed complement to Wikipedia (in the same way as, for example, Wikisource or Wiktionary, are), it's also like a part of Wikipedia, since most images (and other media files) shown in Wikipedia are hosted in Commons. So Commons is, in some way, also part of WMF's flagship project. MGeog2022 (talk) 18:46, 15 February 2026 (UTC)Reply
This might be heretical, but I think Commons is more important than Wikipedia. Of course, having articles about mnay topics is important (especially in time of fake news), but having a free media archive, that covers anything from street imagery, heritage images, history, diagrams, media of historic events, animations of ancient cities, 3D models of mathematic conceptions, etc. in times of dying digital media has an importance that can never be appreciated appropiately. And the changes in media creation is obvious. When I see the first media on Commons, it often is some 640x480 photography of old train stations from 2005. Now we have UHD videos, 3D models, interactive SVGs, E-Books, 360 panoramics and hopefully much more, soon. Interesting time to be alive, right? So or so, we need to have a fast reacting archive, to allow more media formats, and to combat with the larger getting resources to be archived. We may have yearly growths of >1 PB soon, so we better be prepared --PantheraLeo1359531 😺 (talk) 19:07, 15 February 2026 (UTC)Reply
@PantheraLeo1359531, yes, Commons is (or at least has the potential to be) some kind of mini-Internet Archive (excluding, of course, the web archive and non-free media). Wikipedia is also very important as the vertebrating core, the entry point for every topic. From there, you can get more details from Commons (or, for example, Wikisource, if it's about a book).
I think the true sum of all human knowledge is Internet Archive. But it's too big, and content is distributed among many different sites, many of them untrustable (well, that's just the Internet, not Internet Archive's fault :-D). Wikipedia is the summary of all human knowledge, the starting point. Further knowledge is in Wikimedia sister projects (for example, the Commons category about an article's topic), and even more (references and external links, especially for non freely licensed content) is at Internet Archive. When you start from Wikipedia, most meaningful knowledge about any topic should be only a few clicks away.
Thinking about Internet Archive, I would say that its Wikipedia pages (including the parts of Commons that are archived, Wikisource, etc) are like an index for the most important parts of the full archive, the ones that allow to have access to the basic summary of all human culture. MGeog2022 (talk) 19:27, 15 February 2026 (UTC)Reply
@PantheraLeo1359531, thanks for the info, it can be useful to find the cities alphabetically, but, as with the date categories, I fear that I will be a bit "lazy" when I'm doing the uploads (there is enough work to do already).
@PantheraLeo1359531, thanks for the info, I was about to use it just now (you have saved me from a small scare for sure), so I'll wait for a better moment.
Latest comment: 2 months ago8 comments2 people in discussion
I have news: We now have 100 TB new in 2026, and Commons soon has one petabyte of media. Yesterday, I attended a meeting where we talked about archiving of local media and creating them. One attendant told us that our local newspaper company allegedly deleted its photographic archive around 2021 because of economic and legal issues (only limited right of use). The women who had to do that had her heart bled. Some volunteers are planning to acquire photo sets by local photographers for the local town archive, at least. But I experience being the only one or one of few people who take photographs of demonstrations, so this is a bummer. All the best! --PantheraLeo1359531 😺 (talk) 17:39, 25 February 2026 (UTC)Reply
@PantheraLeo1359531, our local newspaper company allegedly deleted its photographic archive around 2021 because of economic and legal issues (only limited right of use): this is really terrible and absurd. I don't fully understand the legal issues (the newspaper company is supposed to own all rights over its own newspapers!), but I don't think the cost of keeping 2 or 3 copies (or even one, if it's in analog form) would be so big. Fortunately, sometimes, a local newspaper is helped financially by the city government to digitize its archive and make it available to the public at no cost (of course, all content is still copyrighted and non-freely licensed). Internet Archive is also working to preserve local newspapers.
Sadly, newspaper archiving can't have a place in Commons, with the exception very, very old newspapers, or newspapers from the USA up to 1931 (fixed-term copyright expiration is a great thing, but I believe it only exists in the USA). MGeog2022 (talk) 21:22, 25 February 2026 (UTC)Reply
I think it is because the newspaper acquired only limited publishing rights from the photographers they put under contract. Several times, local people asked them if they can use some of their pictures, but they are not allowed to give them away. Afaik, our local archive gets one physical copy of every newspaper. But yes, the "Frankenpost" (our local newspaper) was established after the second world war. --PantheraLeo1359531 😺 (talk) 09:47, 26 February 2026 (UTC)Reply
@PantheraLeo1359531, I understand: maybe the photos were licensed to be included in the newspaper in low resolution (and, decades ago, black and white) only, so the newspaper company had no rights over the original ones. Just a fateful combination.
Hi! You probably will retry the failed Mapillary uploads once the first upload process is done. Some images failed because there was an attempt to upload the files while Commons was in a read-only state --PantheraLeo1359531 😺 (talk) 18:09, 5 March 2026 (UTC)Reply
@PantheraLeo1359531, thanks for the info! I already retried: my 2 batches from 17:47 and 17:48 are the retries for the failed uploads. I knew about the problem in the very moment it happened: I was trying to create categories for more Curator uploads when I saw the message. A bit frustrating. I'll try to do the new uploads tomorrow :-) MGeog2022 (talk) 20:13, 5 March 2026 (UTC)Reply
@PantheraLeo1359531, thanks! Just a detail: the text on top is shown in German. Probably, you didn't notice anything strange when looking at it (that is: I understand the full page without problem), but maybe it's not what is expected in this kind of page ;-)
Ok, I will fix the translation :3. https://quarry.wmcloud.org/query/102963 reveals the files until date X. May be interesting, too. I think the Mapillary numbers will increase in the future. Maybe we can sort of enter a greater automatization process. I myself experimented with Mapillary, but it's a bit complicated :D --PantheraLeo1359531 😺 (talk) 14:27, 10 March 2026 (UTC)Reply
@PantheraLeo1359531, I don't think all of them will be in Commons in the medium term, we will have to wait for the time when storage is so good that we have a petabyte disk in our computer :-D. And there are sequences with really bad quality, and very bad images in some good sequences also (I had to nominate some of them for deletion just after upload; by the way, with this, I noticed how slow some deletion requests can be, it seems that nobody processed them after, maybe, a month). MGeog2022 (talk) 13:08, 17 March 2026 (UTC)Reply
@PantheraLeo1359531, thanks, but all my queues remain exactly as they were: no progress at all :-(. If it's not possible to solve it, I'll cancel them and then launch them again, when I have time (maybe tomorrow). MGeog2022 (talk) 19:17, 20 March 2026 (UTC)Reply
The batches 1294–1296 is queued in total, 1290 (209/312), 1291 (5/36) and 1293 (157/284) are paused. Hmm, I would wait until tomorrow, and when no change has happened, than we can report the issue. --PantheraLeo1359531 😺 (talk) 19:23, 20 March 2026 (UTC)Reply
@PantheraLeo1359531: nothing has changed. Can you report it, please? I'll cancel the 3 batches that are 100% queued, and launch them again: I don't want their categories to be empty for several days (they could be nominated for deletion). But the other files that remain queued in partially processed queues, should resume once the problem is fixed. MGeog2022 (talk) 16:08, 21 March 2026 (UTC)Reply
I cancelled all 3, and launched one of them again, but nothing is being uploaded: unless we returned to non-parallel processing (there is a batch launched by you, also in execution), it seems the problem was not fixed for me (and I suppose the same for all users that meet certain conditions) at all, for whatever reason. I'll also launch the other 2 batches, but I'm not optimistic. MGeog2022 (talk) 16:16, 21 March 2026 (UTC)Reply
I am working on my photographs of prime minister Söder and federal minister Bas the next time, but I will track the changes. But I am not worried at all, DaxServer is a very diligent worker here ;) --PantheraLeo1359531 😺 (talk) 16:59, 21 March 2026 (UTC)Reply
@PantheraLeo1359531, about the first petabyte of files in Commons: that would be three of these disks. The text dumps of all WMF wikis (including Commons) with full revision history would fit in one of such disks. Ten years from now, when writing speeds improve to the point to make it feasible to write all this data to 5D optical data storage, maybe it will be ten disks in all. And Internet Archive maybe will be one thousand disks. I hope in 10-20 years several copies of both WMF and Internet Archive are written to this storage technology (for now, a copy of a snapshot of English Wikipedia at some random instant already has). I think the most important technological revolution happening this decade is 5D optical data storage (even more than generative AI), because, for the first time in history, there is a way to store everything that needs to be kept and preserved, in a medium that can last forever, without any human intervention, and in a rather small volume. MGeog2022 (talk) 11:44, 2 April 2026 (UTC)Reply
I remember hearing about this disk since maybe 2018, and wonder, when it will be mainstream :D. But I am sure, we will see something like this in the near future. The ZDF (German broadcaster) also uses LTO tapes and stores more than 60 PB in 2020. The demand for beyond-Petabyte storage is huge, and as data is essential for human life, this will be the way to be. I can remember people complaing on Commons why this image file is 60 or 100 MiB large. Oh well, we will handle data sizes far above 1 PB soon, and the degree of automated user-generated content will grow fast. Even for people like you and me, it is easier than ever to generate much content without doing anything. So, yes, I think a lot people will discuss how our storage of tomorrow look like. I hope, the 5D disks will be usable for anyone soon, so we can leave the path of unwanted (digital) data loss. :) --PantheraLeo1359531 😺 (talk) 14:14, 2 April 2026 (UTC)Reply
@PantheraLeo1359531, the difference from 2018 is that English Wikipedia (or at least, a random snapshot of it, without edit history, so it can contain vandalized articles without any non-vandalized version) is already stored there. Sow now it's only a matter of time: let's wait for cheaper costs and bigger writing speeds, so writing a full 360 TB disk doesn't take 2 years as it takes now, and my dreams will come true ;-) MGeog2022 (talk) 14:19, 2 April 2026 (UTC)Reply
@PantheraLeo1359531: unrelated to this: there were technical problems with Commons databases earlier today: among other things, Curator uploads were failing. There are several hundreds of failed uploads in 2 of your batches (the same happened with 3 of my batches, but I retried them later, but with your batches I can't do the same). MGeog2022 (talk) 15:01, 2 April 2026 (UTC)Reply
Latest comment: 2 months ago11 comments2 people in discussion
I got to know "geogroup" some days ago. If you add it to a Mapillary sequence category, you can see the trail on a map, where the photographs were taken. It is very useful and can be added easily. When editing a page, there is a button on the bottom with Geogroup|level=1 that adds a box with the link to the map :). --PantheraLeo1359531 😺 (talk) 09:43, 28 March 2026 (UTC)Reply
@PantheraLeo1359531, thanks, that's very good. There are a lot of Mapillary sequence categories by now, but maybe some automation tool can apply it to all of them. And from then on, we can use it in all new sequence categories (it's very good to be able to see where the sequence and all its images are on the map). MGeog2022 (talk) 10:46, 28 March 2026 (UTC)Reply
Well, if in the future we have automated sequence category creation, it would make sense that Geogroup|level=1 would also be automatically added to them when they are created. MGeog2022 (talk) 11:20, 28 March 2026 (UTC)Reply
By the way, is there any way to filter panoramics only in the Mapillary web? I couldn't find it, but sometimes it's really hard to find them in some cities in countries without freedom of panorama (other times it's directly impossible, but this doesn't mean that they don't exist). This is leaving most of Africa out of Commons images from Mapillary: very few African countries have freedom of panorama, and in those that haven't, there are really, really few panoramic images. MGeog2022 (talk) 11:30, 28 March 2026 (UTC)Reply
Latest comment: 2 months ago6 comments3 people in discussion
★Your nomination has been reviewed and promoted★ Congratulations! The image you nominated was reviewed and has now been promoted as a valued image. It is considered to be the most valued image on Commons within the scope: Category:Maps of Vatican City. If you would like to nominate another image, please do so at Valued images candidates.
★Your nomination has been reviewed and promoted★ Congratulations! The image you nominated was reviewed and has now been promoted as a valued image. It is considered to be the most valued image on Commons within the scope: Category:Maps of Monaco. If you would like to nominate another image, please do so at Valued images candidates.
This is a deletion request for the community to discuss whether the nominated page should be kept or deleted. Please voice your opinion in the linked request above. Thank you very much!
If you created this file, please note that the fact that it has been proposed for deletion does not necessarily mean that we do not value your kind contribution. It simply means that one person believes that there is some specific problem with it, such as a copyright issue. Please see Commons:But it's my own work! for a guide on how to address these issues.
Please remember to respond to and – if appropriate – contradict the arguments supporting deletion. Arguments which focus on the nominator will not affect the result of the nomination. Thank you!
@Albinfo, thanks for the suggestion, yes, before importing them from Mapillary, I try to select the best sequences that I see for a certain place (for example, a city), and that includes not having annoying things, such as the mount you mention. But, sometimes, I choose to use such sequences, despite the bad impression they may cause, because they have better image quality than other sequences have, or because they are better as a sequence (for example, covering a larger area of the city, or not being too repetitive), or, simply, because they are the only sequence available for a certain area. The mount makes these images worse, but that's very different from them being useless. Of course, for this specific case, it's possible that I made a wrong decision: I made many imports on the same day, so perhaps I was satisfied too soon with the sequence I found. MGeog2022 (talk) 14:27, 12 April 2026 (UTC)Reply
It's interesting to have a whole neighborhood documented on Commons. We can crop, when it's really annoying. But those images that are of no value should be removed. Albinfo (talk) 14:32, 12 April 2026 (UTC)Reply
There are some very bad images, such as the one you nominated for deletion (I exclude them before import when I see them, but it's not possible to review all of them when there are hundreds or thousands of images). The images that have a good image quality but include an object covering a relatively small part of them, should not be considered as having no value. MGeog2022 (talk) 14:35, 12 April 2026 (UTC)Reply
@Albinfo, thanks. I'm doing imports of (mostly) full sequences from all around the world, so I appreciate that users with detailed knowledge about each city do that additional categorization (other users have already done it with other cities as well). But the fact is that I'm working (when I have enough time) mostly in importing, importing and importing, so I haven't done it even for the cities that I do know well, maybe some day :-) MGeog2022 (talk) 14:44, 12 April 2026 (UTC)Reply
@Albinfo, in fact, when I started using Curator, I was worried with blurry images, but I had to think about the sequences as if they were videos: there can be some blurry or repeated frames. They may be nominated for deletion later, no problem, but I can't be obsessed with them, or I won't be able to import sequences. Pinging also @PantheraLeo1359531, since maybe has also had these thoughts and can tell if has reached the same conclusions than me, or not :-) MGeog2022 (talk) 14:54, 12 April 2026 (UTC)Reply
I also had to combat with single low-quality images (also had a black one image in a sequence, hehe :P). I think it depends on how an image is presented. Some are just blurry all over the place, and thus useless. Some may have sharp areas, so they can work as bridge between to clean frames (this can happen, when there is a heavy shake during exposure and the sensor is too slow to scan itself instantly). I usually check manually for faulty images, but this is usually not needed for special HQ contributors like UWRAPID or OSMPLUS_org, as they have professional 360 vehicle cameras and work like Google Street View :). --PantheraLeo1359531 😺 (talk) 15:22, 12 April 2026 (UTC)Reply
@PantheraLeo1359531, yes, those high quality 360º sequences usually have no problem at all, but they don't reach everywhere, and there are many interesting non-360º sequences as well. The matter here is about how to think about the possibility of having very low quality images in the imported sequence: if the sequence takes, for example, 5 pages in Curator, one can have a quick look and unselect the bad images, if there are any. If you are importing 1,000 images, you can take the risk, only: think about it as if you were importing a video that may have some low quality frames, and the same if there are several images that are almost identical. Some user may fix it later by nominating them for deletion, and that's all.
By the way, a suggestion for improving Curator: it would be good to have an optional field for a "sampling rate", that is, to automatically select one of each 2 or 3 consecutive images. There are some sequences that are very good, but have too many images that are almost identical, and thus have no additional value for Commons. MGeog2022 (talk) 19:41, 12 April 2026 (UTC)Reply
@Albinfo, if you are interested in importing more images from Mapillary for Tirana or for any other place, please see here (you can also ask me if you know about a specific sequence or place that you want to have imported, if you are not familiar with Curator tool). MGeog2022 (talk) 14:33, 12 April 2026 (UTC)Reply
I just noticed the Curator tool but haven't tried it yet. I usually only import single – selected – images from Mapillary, not sequences. I'm looking forward to try this tool as Mapillary2Commons isn't working anymore. Happy to have a good alternative to downloading and uploading manually. Albinfo (talk) 14:37, 12 April 2026 (UTC)Reply
It didn't work in my case and I don't think that "Photo from Mapillary" showing up as description twice is really helpful. And it's not so much to change it by hand: [2]Albinfo (talk) 15:24, 13 April 2026 (UTC)Reply
@Albinfo, yes, for a single photo it's easy to do it manually. For a full sequence, it's not a suitable option, and it could be good to have the option to specify a custom text, such as "Street photo of Somecity" (more precise than the default "Photo from Mapillary"). If there is a bug there, it could be opened in Phabricator (@PantheraLeo1359531).
In any case, categorization already helps to better identify what the photo (or 360º view) is about. We already have a good structure of hundreds of thousands of well-categorized images from Mapillary, and most of them have the "Photo from Mapillary" description. MGeog2022 (talk) 12:46, 14 April 2026 (UTC)Reply
Yes, I think having the coordinates and categorization explains almost all at once. IF a the same standard description for 500+ files within a cat is not appropiate, we can change it with one sweep via visual file change :3 --PantheraLeo1359531 😺 (talk) 14:26, 14 April 2026 (UTC)Reply
This is a deletion request for the community to discuss whether the nominated page should be kept or deleted. Please voice your opinion in the linked request above. Thank you very much!
If you created this file, please note that the fact that it has been proposed for deletion does not necessarily mean that we do not value your kind contribution. It simply means that one person believes that there is some specific problem with it, such as a copyright issue. Please see Commons:But it's my own work! for a guide on how to address these issues.
Please remember to respond to and – if appropriate – contradict the arguments supporting deletion. Arguments which focus on the nominator will not affect the result of the nomination. Thank you!
@Albinfo, I'm sorry, but I don't know about the internal working of Curator tool. When the error message only says "query", I can't know more about it than you do. The user who is behind the tool is DaxServer, and PantheraLeo1359531 also knows more than me for sure, but, probably, any eventual internal fix needs to go through DaxServer. You could write to any of them for help, if it fails again and again, and you are very interested in importing that photo to Commons. Another solution would be to manually download the photo and then import it using Upload Wizard, since it's a single file :-) MGeog2022 (talk) 20:15, 24 April 2026 (UTC)Reply
Latest comment: 26 days ago8 comments2 people in discussion
Good evening!
I am bringing some news regarding to Curator. The tool now supports selection patterns like "every 5th image". I also talked with DaxServer about the limits of upload rates, because, depending on what we want to achieve, importing Millions of useful images takes time. So what can Commons serve? Some sequences have very many images, and processing them may take hours. Choose, what to upload. Regards! --PantheraLeo1359531 😺 (talk) 18:01, 6 May 2026 (UTC)Reply
@PantheraLeo1359531, thanks, that was very much needed. Maybe sometimes can make sense to import a 10,000-image sequence in full, if images have enough distance between them, while in a 100-image sequence, 75% of the images can be redundant. So it's a matter of the type of sequence, more than of the number of images.
depending on what we want to achieve, importing Millions of useful images takes time: we have more than 1 million files imported with Curator, by now. Unless we have a way to import a highly representative set of sequences (automatically well-categorized) where we are sure almost all images are good, I'm not sure if it's good to be able to do that. "The best is the enemy of the good": I'm for preservation, but I'm against "diogenes syndrome": wanting to preserve everything can result in preserving nothing, in the end (example: situation of Internet Archive in 2016: had a big earthquake happened in San Francisco by then, it seems the full archive could have been lost). MGeog2022 (talk) 12:48, 7 May 2026 (UTC)Reply
@PantheraLeo1359531, probably, yes (it's just my opinion, of course). In Category:Images from Mapillary uploaded with Curator, it reads: This category contains 1,073,941 files, which is 0.76% of all 141,060,401 files on Wikimedia Commons. So there is still a good margin, for now (and the total number of files does also increase over time).
We already have a good coverage for some countries, but not so good for others. For some of them, the first problem is there is not enough images in Mapillary itself (particularly, panoramic images for many places in countries without freedom of panorama), while for others, there are not contributors with extensive knowledge of the country doing imports with Curator, so only the best-known places have been covered, by users like you and me. I have plans to improve the coverage for some more European countries or the USA, but a good worldwide coverage probably would need from more contributors from several countries (I'm talking about having many countries with the level of coverage that we already have for, for example, Spain, Germany or the United Kingdom).
A separate issue is the case of Ukraine: I suppose it's because of the war, but all Mapillary images are shown as "not available", making Kiev, I believe, the biggest European city we haven't covered here, because it's not possible :-( MGeog2022 (talk) 19:35, 7 May 2026 (UTC)Reply
Yes, Ukraine is totally blocked, at least for viewing. This was confirmed by Mapillary itself. I wonder what a reasonable treshold of imagery may be. I would like to have a discussion about this topic. Let's say (which I am doing right now) we have a walk around a sea that is well-known in the region. Is it relevant? Probably in this case yes, because it has an Wikipedia article. But what about streets and other infrastructure? Depending on the quality of the imagery and significance of the street, what imagery may be relevant? I think imagery of streets may be interesting when we look maybe 70 years in the future. Today, these images look boring because we see them everyday. But then, around 2100, people may laugh at the images because of the techniques used. Cities are clearly relevant, also in suburbs, as you can learn a lot about demography, a country's unique elements (American suburbs may serve different purposes than the German "Vorstadt"). Where to draw the line? This might be difficult to do... I think the biggest problem is that the usage of a file can cover many cases, some of which we cannot imagine (yet). The value of images increases with quality and metadata and proper descriptions... I have some images that could invite to further discussion. What do you think? :) PS: Maybe in 40 years, Mapillary also covers the Moon :P :D.
@PantheraLeo1359531, yes, I understand. Excluding bad quality images, and also redundant ones (including photos taken 2-5 years later at the same place if no significant changes have happened), probably all street photos in populated places have some interest.
In any case, the important point here is WMF's capacity: if it can have, let's say, 10 PB of Commons media with the current setup (2 production copies and 2 backups, all in RAID, making 8 copies in all), no problem to have that. But if the answer is: well, we can have 10 PB, yes, but maybe we can have simple, non-RAID disks, that can be the beginning of the non-preservation of the data in the long term.
If that happens, perhaps a solution could be to have a new type of secondary content, kept on a best-effort basis, while keeping the files that have a special historical (or non-historical) importance with better preservation warranties (that should include, for example, featured/valued/etc images, all files ever used in wikis, or relevant documents of any kind). A subset of the latter ones could also be provided as media dumps for even better preservation warranties, since, with the current total size, I don't believe that full media dumps can ever return.
That is, maybe Commons can take the risk to randomly loose 1 in 1,000 images from Mapillary, but it shouldn't take the risk to randomly loose 1 in 1,000 Commons files, because the lost file could be a very important one. MGeog2022 (talk) 12:24, 8 May 2026 (UTC)Reply
Okay, you make points I totally agree with. First of all, less is better because it is easier to store in redundant manner. To conform with this, we have to figure out what is precious to be preserved. You said, similar images and duplicates are not, populated places are. I try to take the quality issue seriously, so I find sort of a workflow that prefers HQ imagery. Some users contribute with continuous HQ, which makes it easier. I am really excited what the next steps will be in data aggregation. What helped me, was discovering that some little villages in my surroundings have imagery, which could be used for Wikipedia articles. This would be great, having HQ panoramics of any village, town, city, you name it :). I thought it would be cool too, if we would have a complete track of the Trans-Siberian Railway or Route 66. But this is maybe too eager for today :P --PantheraLeo1359531 😺 (talk) 16:19, 8 May 2026 (UTC)Reply
First steps tutorial
Getting help
Goodies, tips, and tricks
Made a mistake?
Done Don't worry, mistakes happen ;). I fixed the categorization with VisualFileChange. I filed a request to automatically add the milliseconds to the title, because sometimes 2 or more images are taken within one second, and they are uploaded to the same file name. The millisecond timestamp makes every filename unique. Greetings --PantheraLeo1359531 😺 (talk) 19:35, 21 January 2026 (UTC)
.svg/20px-Pictogram_voting_comment_(orange).svg.png){kind=small}
_2016-04-08_16H32M10S629_(752448545440522_at_LCy6mUUv1YfDOj0OX6L4kw_with_FP2).jpg)
_2013-10-08_12H56M25S000_(204931611211760_at_vRAkiX3qWk29RUTFqnvwvw).jpg)
_2016-04-08_16H32M10S629_(752448545440522_at_LCy6mUUv1YfDOj0OX6L4kw_with_FP2).jpg)
_2013-10-08_12H56M25S000_(204931611211760_at_vRAkiX3qWk29RUTFqnvwvw).jpg)
_(denisju_osm)_2025-03-22_06H35M32S282.jpg&action=edit&redlink=1){kind=link}
_(denisju_osm)_2025-03-22_06H35M32S282.jpg){kind=link}
![[2]](https://commons.wikimedia.org/w/index.php?title=File:Mapillary_Blloku_Tirana_by_denisju_osm_-_2025-03-22_06H18M03S311_(4449737345253009_of_nZcf2CRPTuK59wxDXOb0tN).jpg&diff=prev&oldid=1196680173){kind=link}
_(denisju_osm)_2025-03-22_06H35M33S317.jpg&action=edit&redlink=1){kind=link}
_(denisju_osm)_2025-03-22_06H35M33S317.jpg){kind=link}
_(neogeomat)_2019-01-26_14H27M02S000.jpg){kind=link}