Daniel Barca

Tag: google

  • Google Photos Takeout uselessness can only be attributed to malice

    Yesterday I was reminded of how impressively useless Google Photos is. During the last few years I’ve become increasingly aware of the importance of choosing services that take their own disappearance into account, or the fact that I may want to stop using them.

    For software services, I try to go for open source alternatives, regardless of wether they are free or not. For example, I’ve chosen Bitwarden as my password manager because it’s built on an open source core, but I pay the premium subscription because of the advantages it provides. I like the peace of mind of knowing I can choose to export all my information and go to another service at any time.

    A few years ago I decided to choose Google Photos as my photo storage service. I happily uploaded decades of photographs into it and was very happy with the result. Google Photos has a great user experience and a very nice UI… unless you want to leave.

    As my photo library grew larger, I started reaching the 200GB limit that I was subscribed to. The price of increasing the storage on Google’s cloud services was just unacceptable for me, as I realized the system wouldn’t scale properly if my library grew even more.

    So I decided to get a NAS system and leave. But leaving Google Photos is not so easy. Due to some European laws, Google is basically obligated to provide some sort of download system so users can download their own data. If we decide to download all our pictures through this system, we get a bunch of compressed files with all our pictures. Next to each picture, Google includes an additional JSON file containing the metadata information for that picture.

    If we want the library to be usable, we need to process the whole thing through custom scripting programs that are found online, and are extremely complex and unsafe for non-programmers.

    There is not a single good reason not to include the metadata information within the picture itself. The only reason for Google to include a file type that only programmers understand is malice: they know a user is trying to exit their platform and are trying to obfuscate the process. I’m sure Google has a thousand excuses for something like this, but a massive company like this spares no resources on the dumbest imaginable details. There is just no way this can be attributed to incompetence or lack of resources, it is a business decision.

    To further make my point, there’s another step in the transition that Google intentionally tries to make not hard, just impossible unless you have a tiny library. Now that we downloaded all our pictures, how do we liberate our existing Google Cloud storage so we can stop our subscription?

    If we decide to stop our data subscription we must delete all our pictures, because if the standard 15GB of storage are full, we risk Google’s most important service to stop working: Gmail. Gmail is often not just important for people, it’s essential. It grants access to every other online service and it’s where we receive all kinds of information like government notifications or bank movements confirmation details.

    Since Gmail is omnipresent we’re pretty much Google’s hostage and they know it. They will use this to force us to keep a subscription fee that we don’t want or need, and thus they make the deletion of the photos library impossible.

    The only way of automatically deleting our photo library is by inserting a very specific piece of code while the Google Photos library is open. Again, just like happens with library processing programs, this is an extremely complex and unsafe procedure for non-programmers, leaving them the only alternative of deleting the pictures selecting them one by one.

    This kind of behavior from Google is why I decided to entirely ditch their whole ecosystem, swapping GMail for Fastmail and my own email domain, and their search engine for Kagi search, and Google Photos and Drive with my own NAS system.

    I’m convinced that Google is engaging in anticompetitive behavior with this, and I’m surprised that no government entity has decided yet to step in and force Google to make this process usable.

    The kind of modifications that need to be made to make the system reasonable are extremely easy to make for a company like Google, and I bet that they actually already have them somewhere in their codebase, just disabled for end users.

    I’m ok with companies making it inconvenient to leave their ecosystem, but it has to be done by the sheer quality of the service itself, not by adding artificial barriers around it. Photo libraries are an extremely sensitive and personal matter, and companies shouldn’t be playing with people’s lives like this.

  • How internet search is changing forever

    Google used to be great. At any time, you were just a few taps away from the answer to any question you can imagine.

    For the last few years, though, Google has become increasingly enshittified. Trying to solve a problem using Google leads to a collection of bullshit articles written automatically by bots and AI.

    When I started interacting with those sites, I thought Google would very shortly bury them in the search results. But for some incomprehensible reason that’s not happened. Only Google knows why they have decided to let their primary product in a state of slow decay. For the last couple years searching has become useless. I find myself appending “Reddit” (or any other specific reputable site) to all my searches if I had any hopes of finding useful information.

    So the solution to a technology that fills the internet with bullshit is another technology capable of filtering it out. In that context I discovered Perplexity.

    Perplexity is what happens when you let ChatGPT search on the internet. Instead of spitting random words that may or may not be true, Perplexity analyzes the question you give them and bases the answer on search results coming from that question.

    The result is an answer that includes references and links to the information it’s based on. It’s able to give reasoned answers like the ones you’d get from ChatGPT, but also give you some confidence on the credibility of its answers, or at least the opportunity to contrast them.

    After a few weeks, I found myself slowly replacing the “thought” of searching something on Google with the one of looking it up on Perplexity. And as a tech-avid user that’s been using Google daily for decades, that’s saying a lot.

    After failing to find what I’m looking for from Google, getting it in a few seconds from Perplexity is mind-bending. And at some point I will just not bother trying with Google anymore.