dosinga

17 ms·

1. 181.▲
   Python Comes to Google Sheets(thenewstack.io)

   5 points by dosinga 3y ago|0 comments
2. 182.▲
   Use Python Inside Google Sheets(neptyne.com)

   7 points by dosinga 3y ago|2 comments
3. 183.▲
   by dosinga 3y ago

   Sonos does a pretty good job still supporting there old hardware. The ZP100 came out 20 years ago and works pretty well today
4. 184.▲
   by dosinga 3y ago

   There was a reason people used to do layout with tables...
5. 185.▲
   High-altitude heist – 500 dollars stolen(bbc.com)

   42 points by dosinga 3y ago|16 comments
6. 186.▲
   by dosinga 3y ago

   Do we need multiple planets for this? Already there are sub-cultures who have more children - see https://en.idi.org.il/articles/25385 for example.
7. 187.▲
   Show HN: Automate Research with a Neptyne Spreadsheet and OpenAI(neptyne.com)

   1 points by dosinga 3y ago|0 comments
8. 188.▲
   by dosinga 4y ago

   Restaurants use faster dishwashers. Those things are a sight to be seen
9. 189.▲
   by dosinga 4y ago

   >quoting issues Did you mean quota issues? As for our own models, it depends on the use case. If you click the code assist and type "Insert a pie chart of the sources of income into B3", the Open AI models do a hard to beat (at
10. 190.▲
    by dosinga 4y ago

    It's a conscious decision. The spreadsheet recalculates itself, the python code doesn't. Recalculations in the spreadsheet are determined by the graph (cell Q3 depends on cell D4 because it mentions it in it's formula like &q
11. 191.▲
    by dosinga 4y ago

    I think it might have been a quote issue actually ... But we should handle failures better, I agree
12. 192.▲
    by dosinga 4y ago

    While we didn't set out to solve those problems in Neptyne, using actual data types in a spreadsheet does solve this sort of problems. Of course we have to be a little compatible, so our dates do come with the 1900 is a leap year bug t
13. 193.▲
    by dosinga 4y ago

    Let us know how that goes and if there is anything we can help with
14. 194.▲
    by dosinga 4y ago

    Thanks! We like it too :)
15. 195.▲
    by dosinga 4y ago

    thanks! that's helpful
16. 196.▲
    by dosinga 4y ago

    Home/End : Oh that feels like a bug. Stop propagate that event already! > The handlers like "on_dropdown_change" should receive an argument for the cell it's coming from They do! If you define a function that does not
17. 197.▲
    by dosinga 4y ago

    Eh, yeah, I think we are experiencing some slow downs all over the place. Sorry about that? It usually should take about the time of a round trip to the server (plus whatever time Python takes to execute the sum, but that should be fast). S
18. 198.▲
    by dosinga 4y ago

    Sorry, not right now. Obviously something we plan to do eventually ...
19. 199.▲
    by dosinga 4y ago

    Agree with making it as much the same where it can be the same. In this case it should work the way it works with the other spreadsheets though. Click a cell and start typing should work and for me just did when I tried. Can you reproduce t
20. 200.▲
    by dosinga 4y ago

    You have a right to be not all that thrilled because of this. Unfortunately you seem to have hit a bug that we sometimes see and haven't been able to pin down. Normally if the network is flaky or the server slow, it should just reconne
21. 201.▲
    by dosinga 4y ago

    Switching tools is hard, that's something we understand. You can import sheets fairly easily of course into Neptyne. Would a two way sync (where a certain region of a Neptyne document and a certain region of a Google Sheet would automa
22. 202.▲
    by dosinga 4y ago

    Maybe? The closest analog in Excel would be to define a VBA function call it from a sheet, go back to the VBA editor, change the definition and see the function be run automatically everywhere with the new code. That does not happen, I just
23. 203.▲
    by dosinga 4y ago

    Right? It looks exactly like what GPT3 would say!
24. 204.▲
    by dosinga 4y ago

    Hmm, Neptyne doesn't support XLWings and that link doesn't resolve. Where did this come from?
25. 205.▲
    by dosinga 4y ago

    Thanks, this is excellent feedback: * The REPL is supposed to do autocomplete where it makes sense. If you enter A1. it should show you a list of things you can do there. That always feels better than asking for it explicitly. Let me know i
26. 206.▲
    by dosinga 4y ago

    Ah, yes, sorry, just Python based slicing. It follows the numpy rules
27. 207.▲
    by dosinga 4y ago

    Nice. Any learnings that we got profit from?
28. 208.▲
    by dosinga 4y ago

    You are correct, that is the slicing syntax. We to carefully interpret the ":" in the right way. For example: sort(l, lambda A:F1(A)) doesn't invoke Sheet1[10:20] just returns the 10 to 20s row of that sheet, which I think is
29. 209.▲
    by dosinga 4y ago

    Sorry, not yet! I suppose we could allow editing the code pane using an external editor, but the spreadsheet would be tricky I think. Or do you have an idea of how that would work?
30. 210.▲
    by dosinga 4y ago

    Oh, thanks. That's very helpful. Schools do seem like a good area to roll this out and in fact we did a pilot with a Swiss university (I know, not in the EU). We should look into this.