Video Platform Crisis

Five years ago, I experimented with screen capturing under OpenSUSE with a Java Applet recording software that randomly crashed in this setup but still left me with what was recorded up to that point, so I uploaded those resulting videos to YouTube. Back then, each video was required to have no more than 15 minutes playtime and there wasn’t a way to request the unlocking of longer or even “unlimited” playtime. YouTube decided about who was eligible for uploading longer videos in an intransparent fashion, so people came up with wild theories about what could be required. I wasn’t affected much as most of my videos were shorter (partly due to the crash behavior described above) and after some time this limitation miraculously was removed from my channel. Later, YouTube dropped that policy completely.

Next, the first major YouTube design update, at least of my remembrance, “happened”. It removed the ability to customize the channel’s background in order to make the site more mobile friendly and only left the header banner to fiddle with. People were pretty upset as there was quite some effort put into those backgrounds, and from todays perspective, it’s difficult to comprehend why a responsive design should not be able to support some kind of dynamic background. The old layout allowed channel owners to convey context, charm and personality, but users eventually compromised with the new, cold look. My channel wasn’t affected much as I didn’t use a custom channel background, but in principle I didn’t like the downgrade as a user of the platform.

Then there was a long period of mediocrity. Good and innovative features were added like the online video editor, livestreaming and the option to submit and therefore work together on closed captions/translations (albeit channel owners are allowed to disable it and it’s not enabled per default), as well as features which are a required minimum like the removal of the character limit on comments or links in comments. Other good features were removed for bogus reasons, like video answers, chronological and tree-hierarchical display of comments, while other features are still missing like the ability to search for comments, export comments or ad-hoc playlists. The forced Google+ integration as a vain attempt to immitate Facebook was annoying, but also easy to ignore.

In 2015, video platform disaster struck. A big YouTuber complained about all the negativity in his comment section as people were spamming it with insults etc., so Google took action against potential haters. The “solution” was to automatically flag suspicious comments and don’t display them to any other visitor except the potential spammer himself while logged in, so he wouldn’t notice that he is writing his comments into the void, presumably continuing such activitiy. After some time with nobody seeing his messages and therefore not responding, he eventually would get demotivated and stop. The problem with this approach is that the algorithm never worked. My comments were repeatedly “shadow banned” (another term is “comment ghosting”) as I usually write long commentary, adjust the text in short succession in order to correct typographic or spelling errors, or in case I add a link after the initial save. If I invest valuable lifetime to state my opinion and cannot be sure that my texts get actually published, there’s no reason to continue doing so, especially as there’s the risk of never getting replies from people I would be interested in talking to. As a publishing platform, such conduct, tricking people into believing that their contributions matter, is a no-go and not acceptable. This is why I abandoned YouTube and only (ab)use it for SEO since then.

So Vimeo became the new home for my videos. It’s much cleaner and had two nice features I liked very much: they allowed videos to be marked as being licensed under CC BY-SA (instead of only CC BY as on YouTube), even despite this information is well hidden from the viewer. The other feature is that they provided a download button (albeit channel owners can disable it). Being a software developer, I don’t believe marketing claims like streaming a video is different from downloading it. Technically it’s the exact same process and while a stream replay might be cancelled earlier and therefore may consume less bandwith, a full download allows offline replay and could save much more bandwith after all. Just think about all the music videos that get streamed over and over again for no other reason than people not having a local copy of them for an offline playlist. For me, the download button is relevant because I want my freely/libre licensed videos to be shared everywhere and archived by multiple parties. The computer has to retrieve the data anyway in order to play it, so it doesn’t matter if this happens via an explicit full download or automatically in the background the browser, internally saving the received data to disk. Vimeo recently decided to remove this convenience feature for all viewers if the channel owner doesn’t pay a subscription fee. As I regard a download as a view, Vimeo as a publishing platform downgraded its publishing, and that’s another unacceptable no-go.

Furthermore, I guess by complaining about the downgrade to Vimeo support, they looked into my channel and suspected it to be a commercial one. Technically I registered a company at the beginning of the year and yes, there was one video linking to a shop seemingly advertising a product, but the shop is not an active one and I never sold a single item there or anywhere else. While I’m not necessarily a non-profit, it’s an attempt to build a social enterprise in it’s early, non-operational stages. I’m fine with paying for the hosting, but I expect the price to be tied to the actual service and not to the arbitrary removal of software features, especially if extortionary practices are used against me and my viewers. Additionally, Vimeo was taking my original video source files hostage and deleted all of them without providing a way to object to their conclusion. They deleted my entire account including all comments and follows, a big time fail in publishing. That’s why I abandoned Vimeo.

I briefly tried Twitch, but uploaded videos don’t get a lot of attention there as the main focus of the site remains on live-streaming. Joining made sense because I’m planning to stream more, but then I discovered that they run ads for the German military (Bundeswehr) before videos and streams, something I’m totally opposed to as a Christian believer and conscientious objector by both, conviction and approval. This is especially the case after I developed anabaptist views. I don’t mind if Twitch promotes violence and destruction which is none of my business and therefore easy to refute, but I never want to contribute to the endorsement of military or their recruiting, so it’s basically the YouTube adpocalypse the other way around.

After that, I moved to Vidme. I liked that they attempted to educate and familiarize viewers with the concept of paying for the production if there’s a demand, but I wondered if Vidme would be able to pull it off. The site had a “verification” process with the goal to determine if an uploader is a “content creator”, which is strange because they demanded rather arbitrary metrics: one needed 50 subscribers but was hardly able to attract any as there were limits on what can be uploaded if one wasn’t “verified” as a creator yet. In my opinion, they put this requirements into place to restrict the amount and size of uploads to YouTubers with a huge audience who switched to Vidme in response to the the adpocalypse and to deter uploaders with a small audience and huge content collection. The latter only cost money for hosting, the former are supposed to earn the site operator some income. I was already suspicious if such policy might have been their pretty serious business need if their endgame wasn’t to be bought up by you know whom. Now, I have videos that are small in size as they’re just screencasts and yet more than 30 minutes playtime, and without the status of being verified, I was prevented from uploading those, while the very existence of those videos proves that I am actually a content creator. I applied for “verification” but they declined the request, so obviously my presence was not appreciated on their site, so I set all my videos to private. Soon thereafter, they announced that they’re shutting down.

A few notes on other, minor video platforms: there’s Dailymotion, but they don’t have users on the site and lack search engine visibility. My impression is that they’re not in the game of building communities around user-generated content but focus on traditional TV productions (remember Clipfish, Sevenload, MyVideo?). Also, the translation of their site into German needs some work. Discoverability is poor as results are polluted by spam video posts advertising PDF downloads and nobody flags them as abuse, as they’re still lacking the option to do so. Their video requirements are: 2GB and 60 minutes maximum per video. There might be hidden additional limits that are not stated on the site: per day and IP address, no more than 96 videos which in total don’t exceed 2 hours of playtime. The uploader is well hidden: After logging in, click on “Settings” under your profile avatar icon on the right in the top menu to find the entry “Upload” in the sidebar. There’s Veoh that finally got rid of Adobe Flash. There’s Myspace accepting video uploads, but they’re not positioned and percieved as a video site and had a massive data scandal, plus failed to handle it well. There are new efforts like, BitChute, BitTubers, Viewly and D.Tube that try to promote decentralized peer-to-peer hosting. D.Tube for a long time required verification via text message and because I don’t have a mobile phone, I couldn’t create an account. Now there’s the option to complete the process just with e-mail and without Steemit. There might be a limit on the amount of videos that can be uploaded based on the current “VP” count which decreases with every new post. There are still many problems with the D.Tube site, let’s see if these get fixed and improved. BitTubers required me to “verify” my account via phone call — which I refused of course, and subsequently they didn’t migrate my old videos I’ve uploaded there to their new BitTubers site, effectively wasting all the time I’ve invested. We-TeVe is worth a look, but not very popular yet. Note that they enforced a crypto currency miner for some time, so without lending a portion of the CPU power for mining, a visitor couldn’t even find out what’s on the site. Furthermore, one can’t upload more than one video per week. At the moment, I upload most of my videos to the various sites – manually, as they usually don’t provide an API for automatic upload, which is too bad for the scarce, valuable lifetime invested into this very dull, repetitive activity. Unfortunately, developing the Twitch Video Uploader (a fairly technical component automating the uploads via their v5 API) was a waste of time, but still, I could imagine to extend the effort into a fully-fledged video management suite, supporting video upload to many sites.

Which leads to new conceptual thinking about online video hosting. Learning from my earlier mistakes and experiences, I don’t see why the acutal video files need to be tied to a specific player and surrounding website any more, leading to demands and dependence on whoever is operating the service. In a better world, the video data could be stored anywhere, be it on your own rented webspace, a gratis/paid file hosting provider like RapidShare, Dropbox or Megaupload, peer-to-peer/blockchain hosting, traditional video websites or on storage offered by institutions or groups that collect, preserve and curate certain types of materials. The software to retrieve, play, extend and accompany the video stream would be little more than an overlay, pulling the video data from many different sources of which some might even be external ones and not under the control of the website operator. Such a solution could feel like what’s relatively common in the enterprise world where companies deliver product presentations, training courses and corporate communication via SaaS partners or self-hosting. A standardized, open protocol could report new publications to all kinds of independent directories, search engines and data projects (or have them pulling the data after registering the instance automatically or manually), some offering general and good discoverability, others serving niches and special purposes, all of them highly customizable as you could start your own on top of common white-label video hosting infrastructure. You could imagine it as the MediaWiki (without stupid Wikitext as unparsable data model) or WordPress software packages embedding video and optionally making it the center of an article or blog post, but in an even more interoperable way and with more software applications on top of them. The goal would be to integrate video hosting and playback into modern hypertext for everyone as promoted by the Heralds of Resource Sharing, Douglas Engelbart (from 37:20), Ted Nelson and David Gelernter. With the advent of information theory, XML, the semantic web and ReST, it’s about time to improve and liberate publishing in networked environments. PeerTube looks like it could be more or less what I have in mind, but I didn’t find time to look into it yet. But beyond mere publishing/hosting, the real work on hypermedia hasn’t even begun yet.

Update: Sebastian Brosch works on PicVid, which focuses on picture hosting at the moment, but considers video hosting as a potential second target. One downside is that the project is licensed under the GPLv3 where it should be AGPLv3 + any later. Online software doesn’t lead to the distribution of the software, only to the transmission of the generated output, so the user wouldn’t be in control of his own computing. The AGPL, in contrast to the GPL, requires software that’s accessible over a network to provide a way to obtain the source code. Now the user gets the option to either use the online service as provided by the trusted or untrusted operator, or to set up the system on his own computer and use it there, or to task a trusted entity with the execution.

Update: Jeff Becker works on sometimes. Seems to be related to

Update: I just found a community of video creators who produce short clips that speak a language of their own. Their uploads are way off-mainstream, often silly, highly incestuous (them talking about themselves, “drama”), sometimes trolling, meme-related, trashy, usually quite creative and always deliberately amateurish, which in part contributes to their charm. There are plenty of sites that cater to this audience, most notably VidLii. VidLii requires videos to be not longer than 20 minutes, not larger than 2 Gigabyte in size and it won’t take more than 8 uploads per day. With those limitations, VidLii hardly can be considered an alternative to “everything goes” video hosters, but as far as I’m concerned, I would just develop an automated uploader that splits the source material into parts of 20 minutes playtime, append the current part number to the title and won’t send more than 8 parts in one go, especially as the individual pieces could be connected together in a playlist. It’s no coincidence that sites like VidLii recreate the early YouTube layout from around 2008 – not only for nostalgia reasons, but also because in the absence of movie studio productions and advertising money, community engagement was more important than the passive consumption by the masses. The old YouTube felt way more personal, at least in retrospective perception. CraftingLord21 reports on this community and the piece about Doppler gives some more insights.

Here’s a list of similar sites: MetaJolt (did allow uploads, but now limited to mere embedding), Upload Stars and ClipMoon. But be warned: not all of them managed to avoid spam bots, while on the other hand they’re at least not filled with questionable content as for example BitChute and PewTube appear to be. Furthermore, keep in mind that those sites might go down at any time or were never serious to begin with (fake sites that make fun of the other ones), so a time investment of manually uploading or commenting there might just as well go down the drain as it might on the big hosting providers. And make sure that you use a new e-mail account and a randomly generated password if you should ever consider to try them out.

Update: It’s no surprise that people aren’t statisfied with general-purpose functionality around video hosting, so they come up with sites specialized for particular niche audiences. One of the many opportunities is education where you find TeacherTube (behind an adblocker-blocker). Or what about a StackOverflow plus video plus a marketplace maybe? Along the lines of a “MediaWiki” around video, but specifically for journalism/debate, product manuals, mapping?

Update: The number of non-western YouTube alternatives is increasing too, with some of them translating their interfaces into English. looks pretty decent, even if one cannot upload more than 15 videos in total, none longer than 60 minutes playtime, none larger than 1 Gigabyte in size. But be warned: in more than a year, this site didn’t generate a single view for me, and after that time, they just deleted my videos without warning/notification.

Update: Similar to Vimeo and Vidme, Rumble tries to establish a different business model for online video, a different one than advertising. Sure, in the digital age, the attempt to maintain artificial scarcity with exclusive licensing in order to immitate print-era business models that don’t work any more doesn’t make a lot of sense, but as Rumble accepts non-exclusive uploads, it’s at least gratis video hosting. Unfortunately I was unable to open the info box on the upload page in order to read the details of the non-exclusive agreement, so please make sure that you check it if you should consider uploading. Also, Rumble is notorious for rejecting some uploads or copyright strike them for no reason, and despite the support helps out with that, it turns out that they simply “forget” to reset the strike count after clearing the claim, which is why I abandoned that site as well.

Update: On the 2018-05-26, I wanted to comment on a video on YouTube and because of experience wrote the comment in a text editor on my local machine to not loose the text again by accidentally navigating away from the site (without recovery or saving the draft as a feature provided by Google/YouTube) or because of session timeout. Now I wanted to copy&paste, which worked previously, but it seems that they devoted their technicians to prevent copy&paste for comments instead of doing some work that’s actually useful. Now, I know of course that I’m in control of the client and can with F12/Firebug try to get my text in, but it turns out that what’s entered into the textarea isn’t what gets submitted, and the text that’s displayed is just the passive rendering, so it can be that they’re recording keystrokes on the keyboard for the input, leaving no way for copy&paste. I’m not going to retype my text, I rather abandon the usage of my account on this site entirely. Furthermore, with the particular channel I tried to comment on, I have the impression that either the owner is deleting my comments anyway or the stupid Google/YouTube algorithm is removing or hiding them (completely, not even comment ghosting so I could still see them when logged in), or another user is abusively flagging them as spam to get them removed, and I have no way of learning what happened there or to get my text back, it’s just gone and any further conversation prevented. So now this last piece of usefulness was also taken away, leaving nothing than a huge pile of videos to be exported in order to be moved to a much better place.

Update: BitView finally got support for HTTPS and can now be considered for uploads not longer than 40 minutes, for which the video quality will suffer seriously. Don’t forget to turn on HTML5 first as this isn’t the default for some strange reason. BitView is made and operated by Jan, who also runs VidLii.

Update: Verond uses an online software package called PlayTube.

Update: On 2018-10-07, YouTube showed me an advertisement for the German military (Bundeswehr), which is reason enough to not even passively use the channel/account any more.

Update: Worldie allows video publishing, too. Don’t forget to click the “Add video” button at the bottom of the upload page after the upload has completed, otherwise the video won’t get published. It’s also mandatory to fill in the metadata fields before the upload, otherwise the “Add video” button might not appear. For some videos, the upload gets stuck at 99% and never completes for a reason that remains unknown.

Update: As Disqus asks for a blank permission for data use (probably as a result of GDPR) and BitChute uses Disqus for the comments, I drop BitChute as my primary video site. It is a stupid idea in the first place to entrust any site the contacts and comments, but what can you do in lack of a better system? It’s also surprising that BitChute manages to host all the videos, but apparently can’t or doesn’t want to afford to host/serve the comments as well, so a dependency on a third-party service is expected to be accepted. I hoped to use Bit.Tube instead, but found out that one cannot reply to existing replies and doesn’t get notifications for replies. Therefore, only We-TeVe currently has a full copy of all of my videos (except stream recordings), so I’ll link to that one going forward.

Update: Vlare launched recently. The maximum upload file size is 5 Gigabyte, the maximum playtime length is 30 minutes. It’s the continuation of Vanillo and merged with Slivr, but unfortunately, they didn’t migrate over my earlier uploads + account/profile, so they won’t get a complete run of my videos, only new uploads.

Update: Viddla launched recently, but it’s in experimental beta, so most things don’t work properly yet.

Update: Ever heard of Niconico from Japan? At the bottom of the website, the language can be changed to English. Previously, accounts on the gratis plan were limited to 12 uploads, none larger than 1.5 Gigabytes in size, but more recently it’s at 40 uploads, each no larger than 3 Gigabytes in size. If not a paying premium member, certain restrictions did apply some time ago like delays in video loading or especially site visitors being unable to watch videos if not logged in, but these seem to be lifted by now.

This text is licensed under the GNU Affero General Public License 3 + any later version and/or under the Creative Commons Attribution-ShareAlike 4.0 International. See the revisions and rendered revisions of how this text developed.


Autor: skreutzer

2 Gedanken zu „Video Platform Crisis“

  1. Awesome! Thanks very much for mentioning Worldie at the end, and glad that you like it. Ironically, it’s running faster on HostGator’s Cloud Hosting than Amazon’s EC2 Instances, yet 10x less in cost 🙂

    It will be fun to make many channels. Now that the Camp Fire is fading and that huge loss with my mom, I am updating the design and have a new designer. It’s going to hopefully look a lot better in the future.

    1. It’s only at the end because I append updates/additions in ascending chronological order of discovery. Exceptions for sites I’m observing more closely, so I add them even if they’re not ready yet. Quite some sites were removed because they disqualified themselves in one way or another at times when I didn’t have change tracking history for the article text yet.

Kommentare sind geschlossen.