numlocked

Hmm, depends what you mean by 'fancy'. There is a 'full screen' view of any given report: https://demo.sqlexplorer.io/50/?fullscreen=1&rows=1000&query...

And it can do PDF export (with a plugin).

Would love to hear what you have in mind?

Cheers! Yes - lots of good options, and you said it well.

Nope - but would love to do it. At the moment you can clone the repo and run start.sh which should work but obviously is not bulletproof like a docker image. Feedback and PRs welcome!

It’s absolutely open source - and completely free for commercial use. That license simply encumbers that specific functionality from resale. I don’t want anyone selling a SaaS version for profit (if someone wants to do that, they can contact me and we can talk about it).

Yep - that makes sense. The Query pane can be collapsed, effectively hiding the SQL from the end user. This is indeed how a number of people use Explorer. But it could certainly be more optimized, in the direction you suggested. I'll think about how this might be improved!

That's amazing! Thanks so much for sharing. Happy to chat if there are specific features or functionality that would be useful. Always looking for more feedback.

Perhaps a bit impolite, but no offense taken. It's a very crowded space and there are a ton of good tools! I work on SQL Explorer simply because I get to make the thing that works best for me.

Redash is very focused on visualization. SQL Explorer is not. It is going more in the direction of in-browser analysis.

Thanks so much! Can you expand on the idea -- what do you mean by faceted search? What would that look like?

You got it! You'll need three values set in settings.py:

https://github.com/explorerhq/django-sql-explorer/blob/64170...

  def user_uploads_enabled():
      return (EXPLORER_USER_UPLOADS_ENABLED and
              EXPLORER_DB_CONNECTIONS_ENABLED and
              S3_BUCKET is not None)

It should be in the docs, but I'll make sure it's more prominent!

My web server is a public s3 bucket :(

Well, except for the webserver with nginx or apache, correctly configured with SSIs enabled. It's not terrible -- but not 'obviously' less complex than using e.g. 11ty (a static site generator) + uploading to an s3 bucket or FTP server. Just trading off complexity in one place for complexity in another.

A sibling (top-level) comment suggest a standard HTML templating (or SSI) language that many static hosts would support. I like that idea quite a bit conceptually, though of course much easier said than done.

I almost agree. The thing that keeps me coming back to static site generators is HTML partials. Having a header.html file that defines my navigation, that I can edit once and include everywhere, is why I pay the complexity tax.

And since I'm already there, I do also like scss + file watching / hot reloading. Though would happily live without it, if there was a 'native' solution to e.g.

   <html>
   <% include './header.html'>
   <body>
   ...
   </body>
   <% include './footer.html'>
   </html>

Being MIT licensed, does that mean that another company could also offer this as a hosted solution? Did you think about encumbering with a license that allowed commercial use, but prohibited resale?

Also, somewhat related, years ago I wrote a very small framework for fan-out of Django-based tasks in Celery. We have been running it in production for years. It doesn't have adoption beyond our company, but I think there are some good ideas in it. Feel free to take a look if it's of interest! https://github.com/groveco/django-sprinklers

That makes sense. Regulating deployment may simply be the only option available -- literally no other mechanic (besides banning releasing models altogether) is on the menu.

Jeremy -- this is interesting and worthwhile. Thank you!

In the same spirit (ignoring the question of whether this sort of attempted regulation is a good idea), I have a question:

Debating release vs. deploy seems a bit like regulating e.g. explosives by saying "you can build the bomb, you just aren't allowed to detonate it". Regulation often addresses the creation of something dangerous, not just the usage of it.

Did you consider an option to somehow push the safety burden into the training phase? E.g. "you cannot train a model such that at any point the following safety criteria are not met." I don't know enough about how the training works to understand whether that's even possible -- but solving it 'upstream' makes more intuitive sense to me than saying "you can build and distribute the dangerous box, but no one is allowed to plug it in".

(Possibly irrelevant disclosure: I worked with Jeremy years ago and he is much smarter than me!)

Ah! I figured it out. User error. I had hit "open file" instead of "open database" and the application decided to open the 30mb sqlite DB as text (e.g. a sql query), including syntax highlighting. Simply letting it grind away for 10 minutes or so allowed it actually open the file, at which point I could close the tab and performance is back to normal. Look forward to experimenting!

Hi - I am trying to run this on an Intel Mac. I got it to launch once, and attempted to open a small (~30mb) sqlite DB, and got a crash. It now hangs/freezes as soon as the application opens (taking over > a full core of CPU e.g. 120% plus utilization in Activity Monitor), even after deleting and re-downloading the .jar file. Any ideas? Would love to check this out!

edit: Apple silicon, which I note because the process type in activity monitor is 'Intel'

thanks much!

Is that documented somewhere? The "contributing" link in the readme also 404s. I would definitely need to understand the licensing and how contributing works before I could consider integrating this (and I would definitely consider it!). Cool stuff.

“At night, I’m my own boss. There’s no better time for programming. And it’s so much fun because I’m making progress.”

Love that! For me, there’s nothing quite like having a programming project you are excited about to get in the flow and have time melt satisfying by.

Any idea why this is not on the front page? 103 points in an hour seems more than sufficient.

I'm seeing this as well. I don't quite understand how it's doing that in the context of LLMs to date being a "next token predictor". It is writing code, then adding more code in the middle.

Preference panes in nearly any application do this. Sensible defaults, with the option to change them.

Very cool site, however...

...my takeaway is a little different than what is in the commentary box (for the year 2017 in particular). The distribution of incomes don't actually look that different, to my eye.

If this is the grand reveal -- showing that childhood heavily influences future financial mobility -- it's not super obvious. I mean, yes, there seem to be a bit of a skew towards low earners in the bottom tranche -- but really it looks like the group that has had some astounding headwinds is kinda sorta doing about the same as the 'no adverse experiences' group? That is amazing as well!

It'd be nice to be able to get to the underlying data more easily, and drill into see the statistical conclusions. The horizontal bands not being of even length doesn't help either.

Edit: I don't think I was correctly taking into account the "no data" group, which makes the skew much more obvious (that the "many adverse experience" group has substantially lower earning power). I wish that the horizontal groups were of the same length, and the "no data" group was simply removed. I think that would make a transformative difference in terms of actually being able to understand this visually and intuitively.

Edit 2: Also how amazing is it that this study got done! The link to the study is very hard to find on this site, and also is wrong. The correct link (I think anyway) is https://www.bls.gov/nls/nlsy97.htm

Cool! How do you deal with irregularly shaped data? For instance if I have two levels of headers, the first of which spans multiple columns via merged cells?

I think a more interesting dichotomy than "productivity" vs "business" is "efficiency" vs "efficacy". Our work culture (and business pop culture) has come to worship at the shrine of efficiency & productivity, but spends relatively little time dwelling on efficacy. What is the impact that only you can drive? The Effective Executive shaped my thinking about this quite a bit, and I'm a better professional for it.

This is the most interesting exchange in my opinion:

Do you think you integrate your Tetris skills into your everyday life?

Tetris has had a major impact on how I think about randomness. For example, it’s very hard for humans to estimate when things will occur, and a big part of that is recognizing that things are actually independent. Like, if there is an event with two outcomes of equal probability and the event has occurred as just one of the outcomes the past ten times, the probability that the next event will have the opposite outcome is still just 50%, not any greater.

About an hour drive from Reykjavík, Iceland.

What is “model negotiation” in this context?

Ooooo, thanks for the recommendation! I will give it a go.