Comment

Comment on lemmy should tots let you post things to your own user profile

To go back to the webmail example, we could have said “no need for hotmail/gmail because Eudora or Pegasus already exist.” “No need to have Google Maps because MapQuest already has a desktop client”.

Yes, we didn’t really need any of these, but the problem with this thinking is that it assumes that the progress of software application is linear and “intelligently designed”, when it reality it much closer to how actual life evolves, by testing many different adaptions and keeping those that make them more fit to their environment.

It doesn’t matter that kbin already have certain functionality if its main developer was a control freak who was holding back its evolution and its users had not trust in him. There were other features that it was lacking (no API, no third-party clients, not easy to deploy, moderation tooling, etc) and still do. We can not just tell someone “what you want is on kbin, use that instead”, because there will be different use-cases that kbin does not fulfill.

Software co-evolves. Lemmy should “steal” from mbin, as it should steal from Pleroma, or Mastodon/PixelFed is now “stealing” things from Bluesky. This is wasteful, but is at least robust.

If software was “intelligently designed”, we will not have any server-side platforms and just have “Generic ActivityPub servers” that can handle the messages being passing around actor inboxes, and we would all be using client-side browsers that are aware of the ActivityPub vocabulary. But this will be like the GNU/Hurd of the Social Web, and saying that server software should have each only have one defining feature is a recipe to have the whole ecosystem ossified.

source

Sort:hotnew top

Kichae@lemmy.ca ⁨1⁩ ⁨year⁩ ago

We can not just tell someone “what you want is on kbin, use that instead”, because there will be different use-cases that kbin does not fulfill.

So instead, it’s “let’s beg Lemmy to fulfill these use cases that it currently does not”. Got it. Makes total sense, and is not internally incoherent at all.

Definitely not just arguing for a monoculture.

source
- rglullis@communick.news ⁨1⁩ ⁨year⁩ ago
  Definitely not arguing for a monoculture. You are overreacting and reading whatever you want, instead of what I’ve actually written.
  
  I’m not saying “people should leave mbin and use only Lemmy as the end-all solution”. I’m saying “those who are already on Lemmy should not be forced to adopt yet-another tool just because some other alternative fulfills one use-case better”.
  
  mbin might make some of what Lemmy does and it makes some of what Mastodon does, but it is not a perfect replacement to neither. There is always a cost to adopt any new piece of software (and I’m not talking about price, here). If some users are happy with it, by all means let them continue using it, and I hope it keeps improving. But to think that is reasonable to tell everyone “Lemmy doesn’t do this, use mbin instead” is like saying “Linux is not good on the Desktop, use Windows instead”.
  
  source
  - Kichae@lemmy.ca ⁨1⁩ ⁨year⁩ ago
    They’re websites. You’re arguing that people shouldn’t use different websites. On the Internet. Which is kind of how the Internet’s been going the last 15 years, and has turned out to be a total disaster.
    
    The idea that the largest game in town should adopt the features of smaller players, rather than users exploring other options because there’s a slight inconvenience to the user just seems, I don’t know, incredibly entitled. It’s also how smaller projects stay invisible and die, leading to a monoculture.
    
    So no, you’re not arguing that “we should have a monoculture!”, you’re just saying “people shouldn’t have to make choices!” which… leads to monoculture. And overwhelmingly supports the status quo.
    
    source
hendrik@palaver.p3x.de ⁨1⁩ ⁨year⁩ ago
I'm not sure about that argument. I mean there are right and wrong tools for a job. There are people constantly trying to drive in a screw with a hammer. They might be better off with a screwdriver. We could also devise a multitool, or not do it. Ultimately, if just the right tool is in front of you, you'd better have a specific reason why not to use it... I can see one general argument, and that's competition is good or more general tools are good. But that's kind if what MBin is...

Not having server-side platforms is a very interesting argument. I mean most users are using smartphone apps anyways... I don't know why we bother with translating everything twice and doing that many server-side things. ActivityPub with it's concept of inboxes and outboxes is kind of designed to run with a minimal server and do most logic and rendering client-side anyways.

source
- rglullis@communick.news ⁨1⁩ ⁨year⁩ ago
  
  Ultimately, if just the right tool is in front of you
  
  That’s the thing. The “better” tool isn’t right in front of me. OP didn’t know about it.
  
  And if they did, they would have to do the whole “which instance to join” dance, again.
  
  And after they joined the instance, they would have to find a mobile app to use. (oh, oh, there isn’t one)
  
  And after they said “fuck it, let’s just use the web UI”, they’d be like "okay, I can follow people from mastodon and I can follow Lemmy groups, but if there is an user that is on Lemmy, it still means that they can only post things to groups. (IOW, incomplete interoperability resulting in functionality silos)
  
  Wouldn’t it be a whole lot simpler/easier if Lemmy had the capability to let the user create posts without referencing any group? Even if it wasn’t the main feature, it could be implemented at the server and documented just enough for those working on alternative clients. The “purists” that don’t care about the functionality would still keep their tools intact, but the others would greatly benefit.
  
  source
  - Kichae@lemmy.ca ⁨1⁩ ⁨year⁩ ago
    
    they would have to do the whole “which instance to join” dance, again.
    
    Oh, this again.
    
    Seriously, you’re now arguing for a monoculture and centralization.
    
    At that point, just go back to Reddit.
    
    source
    rglullis@communick.news ⁨1⁩ ⁨year⁩ ago
    One more of these ridiculous “so you are saying…” comments, and I will start responding to you with Cathy Newman memes.
    
    source
    -> View More Comments
  - Blaze@feddit.org ⁨1⁩ ⁨year⁩ ago
    
    That’s the thing. The “better” tool isn’t right in front of me. OP didn’t know about it.
    
    People mentioned Mbin in this very thread
    
    And if they did, they would have to do the whole “which instance to join” dance, again.
    
    fedia.io
    
    And after they joined the instance, they would have to find a mobile app to use. (oh, oh, there isn’t one)
    
    github.com/jwr1/interstellar
    
    source
  - hendrik@palaver.p3x.de ⁨1⁩ ⁨year⁩ ago
    I agree like 80%. I think with the tools, it is how it is. They don't necessarily owe you anything. I did some renovations lately, and realized (again) that choosing the right tool and method might be essential. We wasted hours and hours in some cases doing some amateur work. And after asking someone how to do it properly, or getting a recommendation to rent the professional tool instead of bothering with the consumer-grade power tools, something that would have taken days, got done in one morning. Same applies to computers in my experience. I learned how to use some tools that just make things a lot easier. Sometimes they automate a boring task. Or make me 100x as fast. And it doesn't really help to complain I was not aware of it's existence. That's how it is. You can't know everything. You'll have to make ends meet with some amateur work then. Or somehow become aware of how to change things to your advantage.
    
    And things like being locked in, or being invested in something can be problematic with software. Sure you don't want to begin all over if you put in quite some effort and labor. Or learned something for a long time and now you have to switch to something else. I believe that's one of the main reasons why people stick with Microsoft Windows. Despite it not being particularly great. Ultimately that's your choice. Either you put in the effort, re-learn a few things and adapt your workflow. And that's somehow worth it to you and you'll start to benefit from it after a while... Or you don't do it.
    
    But all of this is very abstract. And just from the user's perspective. My point mostly applies if you're the user and faced with a fixed situation which you can not change. Of course that does not apply to the software developers. They should listen to the requests of their users and implement features unless there is a specific reason not to do it. And that's where I completely agree with you. It would be great if the software was capable and had a lot of features. I mean there are some limitations in practice, you need someone to invest time to implement it. And feature creep kills projects, you can't add everything... But I think Lemmy could really benefit from some more useful features. And I don't see a reason why they should reject them without a soecific technical reason.
    
    source