Most active commenters

    ←back to thread

    1222 points phantomathkg | 19 comments | | HN request time: 0.214s | source | bottom
    Show context
    _def ◴[] No.44064357[source]
    > Your export file will include links (URLs) of your saved items. The export does not extract the text of saved links. Additionally, the export does not contain tags or highlights.

    boo! without the tags, the links will be mostly useless for me. Every now and then I thought aboyt switching to some self-hosted solution. Should've done it sooner... and I will never trust Mozilla with any service again.

    replies(12): >>44064387 #>>44064622 #>>44064633 #>>44064647 #>>44064701 #>>44064943 #>>44064978 #>>44065001 #>>44065605 #>>44065678 #>>44067559 #>>44068746 #
    1. chimeracoder ◴[] No.44064387[source]
    > and I will never trust Mozilla with any service again.

    Who will you trust? Google? Apple? Microsoft? It's not like any of the other behemoths have a better track record when it comes to long-term maintenance and availability of hosted consumer products. If anything, Mozilla actually has the best track record out of them all when it comes to long-term offerings.

    replies(7): >>44064419 #>>44064421 #>>44064428 #>>44064465 #>>44064522 #>>44064548 #>>44065019 #
    2. walterbell ◴[] No.44064419[source]
    > Who will you trust?

    Your-self-hosted?

    replies(1): >>44064736 #
    3. xnx ◴[] No.44064421[source]
    > Who will you trust? Google?

    Google has very good/complete Takeout data for most of its services.

    4. ummonk ◴[] No.44064465[source]
    What hosted products has Apple dropped? iCloud might stop supporting old devices but that doesn’t stop you from keeping the data on device or accessing via a new device.
    replies(2): >>44064600 #>>44064630 #
    5. thrdbndndn ◴[] No.44064522[source]
    His point is about data portability, not service uptime.

    This export feature is outright bad, worse than the industry standard by a mile. Why wouldn't it include something as basic as tags? It just forces users to write their own scripts, wasting time for everyone involved.

    replies(1): >>44065822 #
    6. Buxato ◴[] No.44064548[source]
    I trust none of that ones, only Google for one of my emails and my Android phone. The track of abandoned services of Mozilla is astonishing.
    7. madeofpalk ◴[] No.44064600[source]
    iTunes Ping https://en.wikipedia.org/wiki/ITunes_Ping. I never trusted Apple after they got rid of that /s

    Apple's problem is they'll often leave products to be stagnant. Existing, but on life support. Like basically all their Mac Apps. A lot of hardware products like this as well, like HomePods.

    They have a raft of iOS apps that seemingly come out of hackathon projects that they release, never update, and then maybe quietly kill off. I thought they killed Clips, but it's still hanging out there...

    replies(1): >>44110489 #
    8. chimeracoder ◴[] No.44064630[source]
    > What hosted products has Apple dropped? iCloud might stop supporting old devices but that doesn’t stop you from keeping the data on device or accessing via a new device.

    Apple is a bit of a weird case because historically they've been a hardware company first and have done very little in the way of consumer services. But they're just as happy as to kill off consumer products if they want to; they just have a more limited selection to start from (which is itself another layer to the "problem" of trying to use them as a replacement - you can't rely on a product they don't offer).

    9. reaperducer ◴[] No.44064736[source]
    Your-self-hosted?

    Only if you're among the .0001% of people who can code it yourself. Otherwise, nothing is different; you're always relying on someone else for their software, feature, security, and compatibility updates.

    replies(4): >>44064782 #>>44065347 #>>44065624 #>>44065759 #
    10. walterbell ◴[] No.44064782{3}[source]
    Are we on Hacker News?
    replies(1): >>44065219 #
    11. rockskon ◴[] No.44065019[source]
    Chasing the ever-cycling list of organizations that haven't yet betrayed trust with regards to privacy. If you're going to go with the "who can you trust" angle then you need to qualify that with a type of product.

    Also?

    An organization's past doesn't dictate their present.

    12. 4ndrewl ◴[] No.44065219{4}[source]
    Yeah, bit weird - surely the population of HN is either going to be

    - people who _can_ code it themselves, or

    - people who believe they can get AI to code it for them

    replies(1): >>44066161 #
    13. homebrewer ◴[] No.44065347{3}[source]
    I'm self-hosting a couple of services that stopped receiving updates (including security updates) about 10 years ago. They still serve my needs perfectly well. They're hidden behind HTTP authentication (with HTTPS) so the internet noise ("AI" shit, web scrapers, shodan, script kiddies) doesn't even know they exist.
    14. jauntywundrkind ◴[] No.44065624{3}[source]
    It's a pity Sandstorm never super took off. Portable serverless with really nice security built in meant really easy to deploy apps; app server just for you or for a group of people was so nice & easy.

    I really hope we someday have self-hosting that isn't as intimidating, that isn't a million different systems all complex in their own way, where there's a base platform with base assumptions and base tools, that let's us manage our self-hosted apps & their data.

    15. bee_rider ◴[] No.44065759{3}[source]
    Code? For self-hosting? I hope not. Most programmers shouldn’t be programming internet-facing stuff. (FWIW I include myself in “most programmers”).

    Get it from your repository. In that case you are trusting somebody like Debian or Redhat. They’ve earned an awful lot of trust.

    16. jamienicol ◴[] No.44065822[source]
    It does include tags
    17. doubled112 ◴[] No.44066161{5}[source]
    I'm in a third group:

    people who can decide if somebody else's code is "good enough" to host it themselves

    replies(1): >>44070773 #
    18. 4ndrewl ◴[] No.44070773{6}[source]
    For my definition of code in this context you're in the first group, as you're doubtless going to have to modify some Dockerfile/.env/config.yaml to point the flapdoodle_cache setting to /usr/local/.this/fnar
    19. ikomrad ◴[] No.44110489{3}[source]
    And here I thought I was the only person who used that service.

    RIP Apple Ping.

    It met the same fate as Facebook Ping.