Hacker News new | past | comments | ask | show | jobs | submit | dewey's comments login

Time of the day, day of the week. Other topics that are popular at the time. There’s a lot of factors and it’s not always the quality of the submission. That’s why the second chance pool exists.

There's still a difference between functional, information-dense design and a missing sense of design or care.

An interface displaying a lot of information, and ways to manipulate data can be functional and information-dense if there's thought put into how to do the layout, usage of colors, icons and other factors (Thinking about power user tools like Logic, Video cutting software, internal company tools or POS software that is used by many, all the time). It might not necessarily look "clean" and beautiful to the uninitiated but it gets the work done effectively.


If you think as yourself as an application and you are surrounding yourself by small service applications that do one specific task...then I could see it make sense.

How is that related to a new YouTube design?

The screenshot section in the README seems to be empty. Would've been interesting to see that. There's many tools that do similar things like https://github.com/ankane/pghero or some tools here: https://gitlab.com/postgres-ai

I think it's probably just a large file, as I remember thinking the same thing but it was loaded when I returned to the browser tab with the readme later on.

Ah, you are right. I'm on a slow connection right now and it's a 22MB gif.

(https://github.com/nexsol-technologies/pgassistant/tree/main...)


> and it's a 22MB gif

Whoa... that's a lot of data for a README! But demos are pretty important, so I guess it's worth it.


A mp4 would be a much better choice instead of a gif and probably ~1.5MB.

    ffmpeg -y -i pgassistant.gif -c:v libx265 -q 55 -tag:v hvc1 -movflags '+faststart' -pix_fmt yuv420p pgassistant.mp4
That takes the GIF down to 1338973 bytes (1.3M) with (to my eyes) little loss of readability.

  __264516 Feb 12 11:44  pgassistant.gif
  22782965 Feb 12 11:46  pgassistant.gif.raw.gif
  _2120322 Feb 12 11:55  pgassistant.gif.av1-20.mp4
  __245780 Feb 12 11:56  pgassistant.gif.av1-55.mp4
wget -O pgassistant.gif.raw.gif 'https://github.com/nexsol-technologies/pgassistant/blob/main...'

ffmpeg -h encoder=libaom-av1

ffmpeg -i pgassistant.gif.raw.gif -c:v libaom-av1 -crf 20 -cpu-used 8 -movflags '+faststart' -pix_fmt yuv420p pgassistant.gif.av1-20.mp4

ffmpeg -i pgassistant.gif.raw.gif -c:v libaom-av1 -crf 55 -cpu-used 8 -movflags '+faststart' -pix_fmt yuv420p pgassistant.gif.av1-55.mp4

A place to start from at least, note the 264516 gif is what's currently on the landing page, with the wget command to grab the raw file.


Is there any benefit in using libaom-av1 over libsvtav1 nowadays?

Licensing, patents (as far as I've heard anyway).

NEARLY everything can use AV1 and you don't need your clients to install a licensed codec if their OS didn't happen to include one. https://caniuse.com/av1 Far more than https://caniuse.com/hevc


Submit a PR?

I'm trying but it turns out that it's impossible to submit a useful PR with a video embed in `README.md` because the only way it works is by uploading the video through the web editor[0] and using a static user-images.githubusercontent.com URL.

[0] Which puts it under your account, obvs., and is therefore not that helpful for a PR.


Maybe just open an issue and provide the ffmpeg command then

Yeah, did that in the end.

Good point.

You can also just do "where true", easier to type.

Not in e.g. MSSQL

Sounds easy, but at this point everyone is trained to solve these captchas and implementing the questions is not a quick thing either on a bigger scale (Translations, cultural differences, bots easily bypassing them etc.). I've used captchas on my sites before because bots were just hammering the login form, checking checkboxes and causing me to rack up email sending bills.

There's also https://www.getclockwise.com, it's pretty similar but I picked that over Reclaim for my work account as they had a better privacy policy.

This sounds like a good idea in a vacuum but the reality is that people (I consider myself a very organized person) will just not be able to follow their calendar as instructed as life doesn't work like that.

Instead of checking if their friend is available...they'll just message them and ask them what they are up to. It's a technical solution to a human nature problem which isn't always the best approach.


> will just not be able to follow their calendar as instructed as life doesn't work like that.

This won’t help people who never consult their calendar.

The benefit of this, and many AI tools, is that it takes the first steps for you. The marketing material about creating great schedules automatically is probably bound to disappoint.

However, it changes the game from creating a schedule from scratch to modifying a schedule until it works. Many people find it is easy to critique and tweak, but aren’t good at taking the first steps.

Those are the people who benefit from tools like this. People who don’t even check calendars at all would need something else.


I’ve never seen the reason I can’t stick to a calendar said in such clear terms.

Exactly that.

Scheduling is a consensus thing and the consensus may or may not be reached until AFTER the event.


Hmm, if the tool had some way of tracking whether work was suggested was done, it could be used in a purely private purely suggestive manner where. Like google maps, it could just update the schedule if you don't follow the original suggestion.

Good idea. And if users tracked how long things took on the app, then the personal AI could better schedule your time by having accurate estimates of how long an assignment might take you.

Some landing page feedback: As Slack is mostly used in a corporate context the first thing people will look at when they add an app to their workspace would be the privacy policy and to see if it's trustworthy. Would maybe be worth it to expand on that a bit?

That's a great point! Privacy is really important to me as well, and I'm strongly against collecting anything beyond what's necessary for the app to function. I'll work on adding a dedicated section to clarify this as soon as possible. Really appreciate you taking the time to share your feedback!

Edit

Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: