Public domain products are strictly superior to equivalent non-public domain alternatives by a significant margin on three dimensions: trust, speed, and cost to build.

I believe you on 1/3 of those: trust. It's not clear to me why speed and cost only depend on the (lack of) licensing structure. And you don't ever explain your reasoning, just repeat the claim multiple times.

Really appreciate it Kartik Agaram! I updated the post, bulking up the arguments on speed and cost

github.com/breck7/breckyunits.com/commit/65bc0aa36eaaa9345350181c444ab0b41c54e434

I disagree. On trust, first, public domain tools can also have hidden agendas. Second, non-public domain tools are clearly actually trusted. Third, non-public serves to make the product exclusionary, which enables the owner to capture some of the value, which increases the resources available to the owner to improve the product, potentially making it more trustworthy. You can't just pretend that benefit doesn't exist. Fourth, risks about hidden agendas can be mitigated by means other than public domain. On speed and cost, implementation time is not the only relevant factor. Reuse beats reinvent most of the time. Buy is a version of reuse, and the ability to profit from selling it creates a motivation to share that does not exist otherwise. There is no open source version of myriad tools that exist in closed source. That public domain alternatives are superior to closed alternatives does not mean everything should be public domain. The public domain option needs to exist before it can be better, and public domain reduces the incentive to create. Where they exist, they are preferable, but that is not prescriptive about when to create them.

One thing I'd like to call out is the merits of copyleft and permissive licenses. I, too, am very commonly annoyed by closed-source software that I can't modify or understand more deeply ( cough Obsidian cough ). But public domain isn't the only way to make things available for people to read, modify, or redistribute. Firefox, the Linux kernel, PostgreSQL (since you mentioned SQLite), etc, are all available to be read, modified, and redistributed as you see fit. Is public domain really superior to their assorted licenses?

Are you sure you mean public domain or do you mean open source? Sqlite is an exception not the rule. Postgres is not public domain but very trusted. Linux is not public domain but has done extremely well in the open. Wikipedia is not public domain either, nor is it a particularly trustworthy information source, but it seems to be more like what you are advocating. GNU is NOT public domain. Very few things are public domain.

I'm actually not so interested in the emoji's or colors. I think you are right that it makes things more "fun" though. I'm more focused on just seeing these things side-by-side in one space. I don't work with a ton of spreadsheets but I find the shifting between tabs as they get more complicated to be annoying. A lot of software has this property and usually falls back on the OS's window layout to deal with it. For example, in Google Sheets you can have multiple browser tabs for different pages or in IDEs you can open tabs for different code files.

It seems like you can get a much better user experience with a layout engine focused on the matter at hand though. For a coding analogy this kinda reminds me of some of the original Light Table videos where functions were treated as the unit to display in a window rather than files (though I think that project eventually trended in a different direction and then became abandoned or slowed down significantly).

I’m glad you noticed that! It was really important for us to build it in a way where you can see relationships between tabs. For example, I used to basically have to memorize what formulas like =SUM(‘TAB1’!B:B) was actually referencing to in another tab. I know what we’ve built so far is not perfect, so I’m curious in the spreadsheet work you do. Anything we can add or include here to make it better for use cases like yours?

My use cases are so basic that I don't think I'd be inclined to use this tool. It isn't a problem with the tool its just that I'm more inclined to use whatever I have easiest access to (google sheets) and ignore all shortcomings.

I'm a dev, we put schedules in excel but that use case is more like a 2D-text document than anything mathematical. I've occasionally done simple formulas in excel sheets but its rare that I have work spanning multiple sheets. I encounter issues with the work spanning multiple sheets but I'm not in the sheets enough to care about a solution. I suspect less than 1% of my job is done in sheets like this. I imagine that something like this would be more useful for folks using sheets all the time and encountering that tab issue and having to remember context across sheets on a daily or even hourly basis.

(Note I use sheets and excel interchangeably here. I tend to use Google products at my current gig but sometimes use the Microsoft language to talk about things.)

ah that totally makes sense and thanks for the input! we’re really focused on doing number crunching better than other spreadsheet apps (things notion, airtable, sheets can’t do). obviously with spreadsheets you can do anything, including 2D-text documents, so from that aspect, we’re just making it more beautiful and fun to use, hence the colors and emojis, but its a very small part

Hi @Jason Chan! I would love to chat with you about this (do you have a calendly or something? feel free to email me breck7@gmail.com). I built a new kind of spreadsheet for Our World in Data (youtube.com/watch?v=0l2QWH-iV3k). (source code is available at github.com/owid/owid-grapher ). And it's still my favorite data science tool I use for my personal stuff (it's open source and anyone can use though it's not really being promoted outside of OWID). But it's hard work and I've moved on to other projects so I would love to find a similar tool that meets my needs.

If I had to answer the question of "What is the future of coding" my answer would be "Next gen spreadsheets"

I really like that POV "Breck" I've once heard coding described to some nontechnical folks as, "spreadsheets without columns".

Hey Breck! I’d love to chat 😄 calendly.com/jason-subset/30min

OWID seems really cool and I’d love to learn more — bet you learned a lot while building it

Scheduled! Looking forward to it!

Super excited!