Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Auto-updates all flood at once #2298

Closed
adamlui opened this issue Dec 14, 2024 · 23 comments
Closed

Auto-updates all flood at once #2298

adamlui opened this issue Dec 14, 2024 · 23 comments

Comments

@adamlui
Copy link

adamlui commented Dec 14, 2024

(Please fill out the issue template with your details)

Expected Behavior

Spread them out a little so tabs don't bombard the user

Actual Behavior

Many tabs open at once, causing slight lag + potentially disrupting workflow

image

Specifications

  • FF 133.0
  • TM 5.3.2
  • OS: Win10

Another drawback is for script authors, this incentivizes users to click "Disable Updates" to penalize authors for doing nothing wrong

@derjanb
Copy link
Member

derjanb commented Dec 14, 2024

Spread them out a little so tabs don't bombard the user

I'd argue that it is better to disrupt the user once with a larger number of tabs than to disrupt them multiple times with a smaller number of tabs.
Also updates usually don't open a new tab, unless the script permissions have changed.

@adamlui
Copy link
Author

adamlui commented Dec 14, 2024

I'd argue that it is better to disrupt the user once with a larger number of tabs than to disrupt them multiple times with a smaller number of tabs.

That's just really bad UX but whatever it's your userbase you stand to turn off

Also updates usually don't open a new tab, unless the script permissions have changed.

This is simply untrue (all those updates were for my authored scripts and I changed no permissions) also every update I've ever received has always opened new tabs

@adamlui
Copy link
Author

adamlui commented Dec 14, 2024

I don't understand how you think lagging a user's browser is better but feel free to close this issue since I cannot talk sense to you

@adamlui
Copy link
Author

adamlui commented Dec 14, 2024

In fact users stand to remove even TM (not just punish me by disable updates for my scripts) for (correctly) assuming the manager hurts browser performance (which it does when 10+ updates flood)

@adamlui
Copy link
Author

adamlui commented Dec 14, 2024

From env info shared by my users during troubleshooting I learned they mostly use VM anyway, was just trying to look out for the minority who still use TM but looks like they're on their own

@adamlui
Copy link
Author

adamlui commented Dec 14, 2024

(I link to managers to install in all my docs and don't place TM high is probably why, but for good reason that it crashes when streaming AI response in major Chromium browsers, so I cannot highly recommend unlike VM)

@derjanb
Copy link
Member

derjanb commented Dec 14, 2024

This is simply untrue (all those updates were for my authored scripts and I changed no permissions) also every update I've ever received has always opened new tabs

Then you probably disabled "Don't ask me for simple script updates".

OK, so closing this as WAI for the moment.

@adamlui
Copy link
Author

adamlui commented Dec 15, 2024

Then you probably disabled "Don't ask me for simple script updates".

It's already checked makes no difference

image

...also it's very peculiar how you typically blame the victims (or other browsers) of your bugs or poor UX decisions for trying to improve your software

OK, so closing this as WAI for the moment.

What is that

@adamlui
Copy link
Author

adamlui commented Dec 15, 2024

Like I'm handing you gold on a platter how to not turn off your users (who will punish me too if you don't appease) but you are so resistant or don't even understand how the default behavior of your own software works for a reason I cannot understand because you stand to benefit if you just humbly heed my advice and don't ruin people's browser experience like your competitors respectfully don't

@adamlui
Copy link
Author

adamlui commented Dec 15, 2024

(obtrusively spamming them as mentioned in #2242 is another indicator + your bizarre responses here that gradually exposes you really don't care or respect their UX)

@adamlui
Copy link
Author

adamlui commented Dec 15, 2024

In fact based on my tracking analytics I'm almost 100% certain my last updates which changed no permissions but merely met a lines changed threshold and applied to 10 of my scripts caused unstarring/uninstalls due to this alarming flood behavior (since users of my scripts often use multiple of them based on their feedback)

@adamlui
Copy link
Author

adamlui commented Dec 15, 2024

Also I recall it always asks even if it's 1 line but even if not, you are ignoring the underlying issue which is if there are a lot of pending updates, it is horribly wrong to flood them at once and potentially lock up their browser vs. spreading it out. There is nothing worse than creating lag and a user of your software knows it's the direct cause, what do you think follows? Uninstalls...

@adamlui
Copy link
Author

adamlui commented Dec 15, 2024

Tampermonkey is hardly the only serious player in the game nowadays and I think you're strangely reasoning from a position of sitting in a former throne like you have nothing to lose when that is far from the case now (i.e. every wrong decision you make -- especially ones that don't protect against but foster uninstalls -- drives existing users who were only loyal like me from longtime familiarity to switch to Pepsi, which newer Greasemonkey users are increasingly preferring at least ~50% if not more based on my own experience gathering feedback)

@adamlui
Copy link
Author

adamlui commented Dec 15, 2024

Also Chrome/FF only counts weekly/daily active users respectively so you actually stand to get dethroned, like my GF scripts users stat only goes up but your Web store users stat can actually go down (or growth drop to slower pace vs. Pepsi or equal pace to smaller brands) and because you had years of head start, the currently inflated stat creates a dangerous sense of safety that is somewhat blinding you to the urgency of this and other UX issues

@adamlui
Copy link
Author

adamlui commented Dec 15, 2024

No monopoly ever went down overnight (except w/ government intervention like Microsoft v. DOJ in the 90s) just reflect on that and what is the current state of userscript managers and choices users have today (not 2013)

@adamlui
Copy link
Author

adamlui commented Dec 15, 2024

Also reflect on how you originally overtook Greasemonkey, really ask yourself why did users switch, then are there TMs of today (not 2013) and are you doing anything to supply new reason to switch (away from not to you)

@derjanb
Copy link
Member

derjanb commented Dec 15, 2024

It's already checked makes no difference

If this option is enabled, then Tampermonkey requires user confirmation in case of:

  • Script version downgrade
  • Script data reset (intentional reset or on re-installation of the same version)
  • New/changed @connect
  • New/changed @include or @match
  • Update URL change
  • Updating a locally modified script (changes will be overwritten)
  • Script modification, which disables further updates
  • No include information (the script is not executed on any page)
  • Modification conflict with another tab
  • Deleted by script hoster
  • Blacklisted

So if you see the confirmation dialog and none of the above applies, it's a bug. Please create an issue with detailed information how to reproduce the issue.
If you feel that one of the above points should not trigger a confirmation dialog, please create an issue and explain why.

As mentioned, I want to keep the current behavior of showing all updates at once. So I'm closing this issue.

@derjanb derjanb closed this as completed Dec 15, 2024
@adamlui
Copy link
Author

adamlui commented Dec 15, 2024

None of those applied (you can visit my https://github.com/adamlui/ai-web-extensions commit history to verify) but why are you making me do the all the work when I'm already helping you out with unpaid labor? You create the issue if you want to improve your code (you already copy many ideas from me for free, like you added GM_getValues mere weeks after completing my #2158 which shared code from a script that did the same thing, even your new marketing albeit overly spammy was clearly influenced by me, show some appreciation it's good karma and do it yourself sir

@derjanb
Copy link
Member

derjanb commented Dec 15, 2024

you already copy many ideas from me for free, like you added GM_getValues

Don't think so: #2045

You create the issue if you want to improve your code

Creating an issue about something that I can't neither read in the code nor reproduce makes no sense.

@adamlui
Copy link
Author

adamlui commented Dec 15, 2024

What does "can't read in the code" mean. You listed criteria for a thing occurring, I explicitly revealed said criteria was not met and thing still occurred, thus it is a bug. GitHub Issues are then created to try to reproduce and solve (so not sure how you cannot reproduce withuot even trying yet prevents you from not being lazy to create Issue yourself instead of force me to do your work unpaid)

@adamlui
Copy link
Author

adamlui commented Dec 15, 2024

Don't think so: #2045

A months old idea (not even yours there too) you lazily set aside for not having elegant solution to detecting multi-val or not until my concise one, then you suddenly release within weeks of seeing mine is coincidence? Get out of here (if you think you're the 1st to steal code/ideas from me so it flies under radar you are quite naive, I don't say anything at first because I'm numb to it, doesn't mean I don't notice every time it happens, just I have more urgent things to do than expose/embarass)

@adamlui
Copy link
Author

adamlui commented Dec 15, 2024

I think I'm going to start removing Tampermonkey from all my docs after I do some recovering from COVID (your unrelenting selfishness simply cannot be rewarded w/ my hard-earned steady traffic)

@adamlui
Copy link
Author

adamlui commented Dec 15, 2024

Even your defense is so awkward and weak (innocent programmer would adamantly deny, guilty one says they don't "think" so like subconsciously you didn't completely convince even yourself yet since it's untrue to "know" so not "think" so)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants