Journal

@SchuminWeb

Archives

Categories

Say hello to Piwigo…

16 minute read

July 27, 2024, 3:45 PM

Beginning on Friday, July 26, you might have noticed that the website looks ever so slightly different.  There is a new camera icon in the header, and a new link at the bottom of the pages called “Photography Portfolio“.  This is for a new photography site that I recently launched:

This new site, formally called “Ben Schumin Photography Portfolio“, is pretty straightforward: it is a new photo gallery that will replace Flickr page as the official location for my photography.  The site is self-hosted, running on the same servers as this site.  It uses Piwigo, which a is free and open-source software package for photo hosting.

The reason that I went the self-hosted route was because I wanted to eliminate my dependency on Flickr for photo hosting.  Over the last year, Flickr has had a few things come up that I have found unacceptable, which led me to make the determination that it was time to move on.  The first thing that happened was a software glitch relating to photo uploading that put me out of business as far as Flickr was concerned for the better part of four months.  What happened was that if I did the upload as I was supposed to do, putting all of the metadata for the images into the uploader, including keywords, albums, and groups, the upload would sometimes go through successfully, or it would hang and drop the entire upload, i.e. it was like Russian roulette, because you didn’t know whether your image upload would go through or hang until you were completely done and had submitted it all for processing.  If it went through, great.  If it didn’t, it would sit there forever and never continue.  Considering that entry of all of the metadata for my images could take a long time because I am so thoroughly detailed with it, there was a lot of effort that was at risk with every single upload batch.  I lost enough upload batches that I contacted Flickr support about it.  There, they just said that there was a software problem that they knew about and were working on, and extended my Flickr Pro subscription by three months to give themselves room to work on it.  They also gave me a workaround for the problem, which involved uploading images with minimal metadata (titles and descriptions only, and no keywords, albums, or groups) through the uploader, and then adding in all of the missing metadata in their photo organizer tool after the upload was complete.  I tried this out when I uploaded a bunch of old photo spheres that I did in 2014 and 2015, and found it to be far more difficult and time consuming to complete than in the uploader, to the point where I decided to just wait it out.  The folks at Flickr also told me on multiple occasions that they would notify me once the glitch was fixed.  Unsurprisingly, they never contacted me back like they promised.  I could also sense some annoyance in their responses when I periodically followed up (about once a month) to find out the status of the fixes.  The way that I ultimately found out that everything was working again was to run several test uploads of content with garbage metadata to simulate the size of a real upload.  I did a big upload of photos, and then I pressed the button and crossed my fingers.  To my surprise, it worked.  I did another big upload, filled it with garbage data, and it once again worked.  The uploader hung a few times during testing, but only after everything had landed properly.  I could live with that.  I emailed Flickr support again, and they confirmed that normal uploading had been restored.  Clearly, that promise that I would be notified when things were working again wasn’t worth the pixels that it was painted with, i.e. thanks for nothing.

The other reason for my turning away from Flickr and bringing my photography portfolio in-house was in response to a major policy change at Flickr.  About a year ago, Flickr introduced a policy in their community guidelines called “give some grace” in regards to photos released under Creative Commons licenses.  They described it this way:

Give some grace.  When you choose to grant permission to your photos under any open license available on Flickr, we ask that you give the reuser a 30-day grace period to fix any possible mistake or misuse of your CC-licensed work with no penalty.  Failure to allow a good faith reuser the opportunity to correct errors is against the intent of the license and not in line with the values of our community, and can result in your account being removed.  If you use CC licenses, please understand that we support and adhere to the strategy for addressing license enforcement described in the Creative Commons’ Statement of Enforcement Principles.

As far as my content goes, I regularly released my work under Creative Commons Attribution-ShareAlike licenses until February 2014, after which I switched to a more traditional all-rights-reserved model for new content.  When I used Creative Commons, I generally used the 2.0 and 3.0 licenses.  I have never used the most recent version of the Creative Commons licenses, which is 4.0.  There is a very significant difference in the 4.0 license compared to the earlier ones.  In the earlier licenses, the license terminates automatically upon any breach.  In other words, it treats the downstream user like an adult.  You want to use the content, great: here are the terms under which it is available for your use.  However, if you don’t follow the terms, you do not enjoy the protection that these licenses provide, and you have to deal with the consequences from that decision to infringe on copyright.  Such consequences could mean a takedown notice sent to a hosting provider under the DMCA, or it could mean a licensing fee through Pixsy.  The 4.0 license, on the other hand, requires that the copyright holder notify the image user of any violation and then provide them a 30-day grace period to fix the problem before taking any other legal remedies.  In other words, it involves a bunch of hand-holding on the part of the copyright holder, and, as that woman from the viral meme said, ain’t nobody got time for that.  As far as I’m concerned, the license deeds spell out the rules clear as day.  If you paid attention to what you were doing, there should be no issue in following the license.  If you choose not to follow the license as spelled out in the Creative Commons deed that is written in plain English, then that’s your decision, and you have to deal with the fallout from that.  4.0 requires that the copyright holder take an educational approach to the situation for someone who couldn’t be bothered to do it right in the first place.  I’ve tried the educational approach in the past, and believe me when I say that it doesn’t work.  Therefore, I wouldn’t touch those Creative Commons 4.0 licenses with a ten-foot pole, because that’s a lot of unnecessary hassle when they already had their opportunity to follow the license, and they chose not to.

My policy about image usage outside of the standard terms of the Creative Commons licenses has always been pretty simple.  If you want to do something that the standard license doesn’t permit, then contact me, and we’ll negotiate terms, which will include a licensing fee.  If you’re going to operate outside of the license and didn’t get permission from me first, then we need to address this oversight on the back end, and that includes payment of a licensing fee, because we’re all adults here, and if you use my images, you play by my rules.  I always appreciate the folks who do follow the rules, and the people who choose not to get to deal with the fallout from their choice.

That’s why what Flickr is doing is particularly problematic.  What they did was to unilaterally force everyone who had uploaded works under Creative Commons licenses to follow the terms of a 4.0 license, regardless of what their actual license is.  Even more problematic is that Flickr itself does not allow users to choose the 4.0 license – or any other specific Creative Commons version – when setting a Creative Commons license for photos.  Here is what Flickr gives you as far as options go:

Flickr's options for Creative Commons licensing

All that they give you is the different variants of the license, and no version number.  When you choose one, it automatically assigns the 2.0 version of the license, which is one of the automatic-termination licenses.  So talk about hypocritical: Flickr is requiring people to conduct themselves like they’re on a 4.0 license, but they only allow their users to pick from 2.0 licenses.  If I wanted to pick a 4.0 license and deal with all of that hand-holding, rest assured that I would do that.  That I have not done that should tell people that I don’t want to be bothered by that, since all of that hand-holding takes up a lot of time that could be spent doing other, more productive things, especially when these people already couldn’t be bothered to follow the license in the first place.  By contrast, a takedown notice or a Pixsy case takes relatively little time to complete, and then I can just move on to other things while the designated processes run their course.  So that’s strike one and two right there.

Strike three came more recently, when I received a warning from Flickr relating to a Pixsy case.  I’m not going to go into any further detail about the case because the case is still ongoing, but this is what Flickr sent me via email:

Hi SchuminWeb, In joining Flickr, you agreed to abide by the Terms of Service and the Flickr Community Guidelines.  Your account was brought to our attention regarding images uploaded with an open license for the purpose of allowing reuse of the content.  Specifically, we have been notified by a third-party that an image they used from your account under a Creative Commons license has resulted in a demand for payment.  The image in question was located here: [redacted]  The Community Guidelines were updated on 31 March 2023 and included the following point regarding members who upload images with a Creative Commons license: “Give some grace.  When you choose to grant permission to your photos under any open license available on Flickr, we ask that you give the re-user a 30-day grace period to fix any possible mistake or misuse of your CC-licensed work with no penalty.  Failure to allow a good faith re-user the opportunity to correct errors is against the intent of the license and not in line with the values of our community, and can result in your account being removed.  If you use CC licenses, please understand that we support and adhere to the strategy for addressing license enforcement described in the Creative Commons’ Statement of Enforcement Principles.”  Please take the time to review the guidelines and make sure you are following them, as your continued use of Flickr constitutes your agreement to our terms.  https://www.flickr.com/help/guidelines  Subsequent notices of such behavior can result in the permanent termination of your account(s) on Flickr.  Regards, Flickr Trust and Safety

That warning message is concerning in a few ways.  First, they took the complainant entirely at their word about the situation, and issued a warning based on that.  In other words, they didn’t investigate one bit.  It’s much like when teachers take another student entirely at their word and punish another kid based on it, i.e. the teachers are fully complicit in enabling bullying behavior.  Considering that they took the complainant entirely at their word and never contacted me to get my side of the story prior to issuing a formal warning, that says something else: they don’t want anyone monetizing any Creative Commons content found on their platform, regardless of any sort of grace period, because they didn’t check to see if a grace period was given.  Additionally, someone who is attempting to weasel out of a situation after they got caught doing something might just lie about it in a complaint if they think that it will get them what they want, and Flickr never checked into the veracity of their claims.  That is unacceptable, especially when I am paying to use the service.  I can’t operate under a system where the people in charge will issue warnings without conducting a full investigation into the matter, especially when the complaint is against paid members, i.e. someone who provides revenue to the company and ultimately affects the service’s bottom line.

My stance on all of this is that Flickr is an image host, and nothing more.  They have no right to tell me what I may and may not pursue with my own work, which is published in other places besides Flickr (and may have been published somewhere else long before it ever came to Flickr), regardless of the license type.  Additionally, the Creative Commons Statement of Enforcement Principles is not a part of the legal code that governs the use of Creative Commons content.  It is ultimately an opinion piece, i.e. it is just a suggestion.  It may have been written by the organization that developed the license, but it is not, itself, the license.

This action on Flickr’s part also tells me something else: by issuing me a formal warning in the way that they did, they have chosen the people who use images wrongly over the people who create the content, and have also chosen those same people over the people who help fund their service.  That’s a gutsy move right there, because they’re choosing their principles over their revenue, and I can’t imagine that will work out well for them.  I am reminded of what it says in Tom Lehrer‘s song “Selling Out“, specifically, “I’ve always found ideals don’t take the place of meals.  That’s how it is and how it will always be.”  In other words, at the end of the day, people have to eat, and you need money in order to do that.  That they are willing to alienate their own paying customers over their alleged principles says a lot, and if they’re willing to take that gamble, well, that’s their prerogative.

I actually miss the days when Flickr was operated by Yahoo, because Yahoo saw Flickr as one of many different online properties that it owned, and let it do what it did.  It wasn’t owned by photography people, but rather, it was owned by someone who treated it as a business, and as such, nobody in charge was trying to make a statement about anything with it.  You uploaded your photos, and that was that.  Flickr itself was pretty neutral.  Then SmugMug came in and bought it from Yahoo back in 2018, and started messing things up from there.  One thing that they did was to heavily push pro accounts.  I’m not going to fault them for that, because while Yahoo may have tons of money, clearly, SmugMug isn’t as well-heeled as Yahoo, as they began to impose significant limitations on free accounts as a way to get more pro accounts.  I admit that it worked for me: I was committed to the use of Flickr, so it only made sense to support the service.  So I got a paid account.  In 2019, they partnered with Pixsy for copyright enforcement, offering a Pixsy subscription as a perk for Flickr Pro users.  By that time, I was already using a higher service tier than was offered there based on my own needs, so I wasn’t able to make use of it, but I appreciated its existence, because it told me that the platform supported the rights of photographers to protect the unauthorized use of their work.  That partnership quietly ended in early 2022.  Flickr started to become more hostile towards photographers after that, and thus I knew that it was time to go.

As far as alternatives to Flickr went, I did my research when determining where I was going to land.  I was looking for a few different things.  For one thing, I was looking for a host that was supportive of my rights in my own work, under the terms that I offered them.  In other words, you are a host and nothing more, i.e. no meddling around in my business.  I wanted a platform that supported all of the metadata that I used, i.e. descriptions, keywords, geotags, categorization, and the like.  I also wanted something where I could directly port the content over to the new solution from Flickr.  After all, at the time that I chose to switch photo hosts, I had 16,000-some photos published on Flickr.  I didn’t want to have to manually migrate 16,000+ individual photos to a new platform.  That would take months to do, and I didn’t want to do that.

I looked at quite a few different services and software packages.  As far as hosted services went, I checked out SmugMug, 500px, PhotoShelter, PhotoDeck, and Zenfolio.  For software packages, I tested out WordPress, Piwigo, Lychee, and Zenphoto.

As far as hosted services went, SmugMug seemed like it was a pretty solid service that might have also been able to take over the stuff that I do on Fine Art America, i.e. some platform consolidation.  However, SmugMug had one major drawback to it: they owned Flickr.  It would be like saying that someone didn’t approve of the policies and practices at Sam’s Club, so they quit shopping at Sam’s Club and started shopping at Walmart instead.  Nothing really changes much with that move, because they’re just moving from one arm of the company to another.  I was moving away from Flickr because I didn’t agree with their policies.  What’s to say that SmugMug, being under the same ownership, wouldn’t begin to enforce the same things on their users as well?  And then I’d be back at square one, since at the end of the day, I just want to be left alone to do my thing.  So SmugMug was out because of that.  I dismissed the other hosted platforms pretty quickly as well, because they cost a lot and they would have required manually transferring all 16,000+ images over, which I didn’t want to do.

Based on all of that, it very quickly became clear to me that self-hosted was going to be a better solution.  I already had the server space, and I had enough know-how to be able to run online software packages, having run various prepackaged software solutions for Schumin Web over the years, both for testing and production purposes.  For instance, this site has been running on a self-hosted instance of WordPress for over twelve years now, and it works very well, requiring minimal effort to keep up to date.  Based on my familiarity with WordPress, I started out with a test instance of WordPress tailored for photography, installing photo gallery plugins, and using a theme intended for photography.  I had high hopes that WordPress would be a good match, but unfortunately, it wasn’t, as I was soon reminded that WordPress is a blogging platform at its core, and that the photo gallery plugins that were available were too small-scale for what I wanted to do.  I got the sense that if I were doing photoblogging, it would probably work pretty well.  However, I already have Schumin Web for photoblogging, where I do that and so much more.  That was disappointing, but that’s the way it goes sometimes, that a platform that works well for one application doesn’t work as well for a different application.

When I tested Lychee, I quickly ran into problems there, as it required a different sort of configuration than I typically used, with some stuff hosted in public folders and others in non-public folders, and it ended up requiring the assistance of DreamHost tech support staff to get properly set up.  So that was a strike against them right out of the gate.  Once I did get it running, it worked well, but it didn’t support all of the metadata that I wanted.  I felt like it supported the photos with a title and short description, but that was it.  I couldn’t see myself optimized the way that I wanted with Lychee.  There was also no way to automatically port over content, so that would have required a massive effort to port the content over manually.  Having to port content over manually isn’t the worst thing in the world, but if I’m going to do it, the solution has to be worth it, and Lychee was not that.  Zenphoto was a little bit better than Lychee in a few ways, but I didn’t really like the way it came together, so its period under consideration was fairly short.

Piwigo seemed to be the best option overall.  I wanted to have a site that was like Flickr without being Flickr, and Piwigo fit that, with good organization of photos, support for keywords, geotagging, as well as plugins that provided all of the functionality that I wanted that the software didn’t provide straight out of the box.  One of those plugins contained exactly what I wanted, too: an import tool using the Flickr API to automatically port content from Flickr to Piwigo.  I ran that in my test instance, and it worked very well, with the photos’ landing in my Piwigo installation, metadata and all.  Very good.  Additionally, Piwigo was very portable, as I could download the whole website locally, upload it somewhere else, and it would continue running like nothing ever happened because it didn’t rely on absolute links.  In other words, we had a winner.  It took me about two weeks to build the new Piwigo site, and that involved porting the content over in batches, and then organizing it all as it came in.  Piwigo supports nested photo albums in a hierarchical arrangement, while Flickr only supports a flat categorization scheme, so I was going to organize photos a bit differently than I did on Flickr.  I think that it’s a better method of organization overall.

The only thing that I am not a big fan of with Piwigo is the theming.  While the software itself is pretty robust and does its job very well, the themes that it offers for the front end don’t look nearly as good as the site’s back end.  Even the best theme that I could find, which I ultimately went with, looks kind of clunky, and it makes things look like they’re kind of shoehorned together.  So yes, I feel like the interface is kind of rough, but hopefully I can fix that down the line as I learn more about how Piwigo works and how the theming works.  Additionally, Piwigo was clearly designed by Europeans, using UK spellings of some words and using European-style date notations, where the day comes before the month.  I have not figured out how to localize that to use American-style linguistic conventions, but when I do figure out how to do that, I will change it.  Likewise, once I figure out how to design my own Piwigo theme, I will do that as well, because Piwigo is a really good system, but it just needs to be skinned properly, and the documentation on how to do a lot of this is kind of lacking.  I’ll figure it out eventually, though.

Also, here’s one benefit to self-hosting that I greatly appreciate: being able to go directly into the database and do a lot of stuff that way.  Even the best-designed back end interfaces have their drawbacks, and Piwigo is no exception.  When it came to entering descriptions for a lot of categories, it was easier to go into the database, run a query, and do it that way, entering in a whole lot of descriptions at once.  You don’t get to do that on a hosted platform.  While there is also a hosted version of Piwigo, that database access proved to be invaluable, and that, along with the API (plus hosted Piwigo is a bit pricey), really put me over the top when it came to going self-hosted.

Then there’s one thing that I imagine that many people are surprised about: this photo gallery is located at a new domain name, benschumin.com, rather than being a subdomain of schuminweb.com like I typically do.  I had something of an internal argument with myself about this for a while, before settling on the separate domain.  The idea is that these sites serve very different roles, and thus it shouldn’t be a one-size-fits-all there, either, just like how WordPress wasn’t the one-size-fits-all solution.  Schumin Web, as I articulated in 2021, is ultimately about storytelling.  On this site, I am sharing my stories about that thing that we call life, and it takes the name The Schumin Web for that purpose.  But the photography portfolio site is not for storytelling.  It is there as a gallery of various photography that I have done, with an eye towards displaying individual works with less overall context.  My Flickr was that way, and that was always marketed under “Ben Schumin”, with plenty of links over to the stories on Schumin Web, but was not, itself, Schumin Web.  Ultimately, the struggle was in choosing whether or not to go with the separate branding that I had been using based on how Flickr was set up, vs. folding it into my existing Schumin Web brand.  I felt like using benschumin.com signaled “grown-up photographer”, while branding it under Schumin Web, though well-established as a personal brand, felt like an imperfect fit.  Also worth noting that no decision here would have cost me any additional money, because I had owned the benschumin.com domain name for many years, with the idea of potentially putting a resume or something on there if I ever needed to, but ultimately choosing to redirect it to Schumin Web and calling it a day.  So this was just a repurposing of an existing property.

So my photography portfolio is now in-house.  I admit that there’s a lot that I still need to do, like changing links to Flickr to the portfolio site.  That will take a while for me to locate all of those links and change them, both on Schumin Web and on the portfolio site.  On the photo site, any photo descriptions that reference alternate versions of the same image currently reference the alternates on Flickr instead of in-house.  Likewise, on Schumin Web, a lot of times when I reference a loose photo on here, it points to Flickr.  Those will all need to be changed.  It’s doable, but it’s going to take a while to do, but it also wasn’t 100% necessary for launch.

And that’s the new status quo, I suppose.  Flickr is no longer a place where I particularly want to be, and it goes to show once again that if you want something done right, you need to do it yourself.