Comment by oefrha | Hacker News Reader

Comment by oefrha | Hacker News Reader

Parent Full thread

The original author seemed to talk a lot about funding development/maintenance, so I got curious about what the hell needs to be maintained. I cloned the https://github.com/t3dotgg/vsc-material-but-i-wont-sue-you repo and had a look. Here's a LoC summary:

  ===============================================================================
   Language            Files        Lines         Code     Comments       Blanks
  ===============================================================================
   CSS                     2          142          119            0           23
   TypeScript             32         2026         1650          243          133
  -------------------------------------------------------------------------------
   HTML                    2           59           49            1            9
   |- JavaScript           2            2            2            0            0
   (Total)                             61           51            1            9
  ===============================================================================
   Total                  36         2227         1818          244          165
  ===============================================================================

Among those, 622 lines of TS are hex color definitions for variants in scripts/generator/settings/specific. Most of the rest seems pretty boilerplatey, e.g. look at the 599 lines in scripts/generator/color-set.ts.

So the question remains: what the hell is there to maintain (that takes more than a couple minutes every $godknowshowlong)? I've published and maintained waaaaay more substantial open source projects for years without expectation of any financial contribution.

There's nothing wrong with building proprietary software of a couple of thousand lines of code, including themes. And people should be able to ask for money in exchange for their work.

What's wrong is the bait and switch, as these projects end up being popular because of their FOSS nature.

He had raised about $7.6k total funding using opencollective.

https://opencollective.com/material-theme

that's pretty good, especially for a vscode theme.

Pretty good for a material design theme. The most boring, over used yet ugly and style-over-function design.

Wild.

> There's nothing wrong with building proprietary software of a couple of thousand lines of code, including themes. And people should be able to ask for money in exchange for their work.

The issue is not someone wanting to be financially rewarded for work, however small. That's completely different from saying you need money to "maintain" what is essentially configuration for colors. That's a deceptive use of that word.

Let's call this what it is: a grifter asked people to pay him for the privilege of hacking them.

"What's wrong is the bait and switch,[..]"

Morally wrong, legally not so much. If it is under a permissive license (and it was MIT originally as others have pointed out) you can always cut a proprietary version.

That doesn't take away the right to use the permissively licensed code of course.

> Morally wrong, legally not so much

Of course, but then, the legality of it is irrelevant.

Free markets work because consumers can exercise their freedom to choose, and in such instances it's easy, given that the reason for why many such projects see wide adoption is their FOSS nature.

> Free markets work because consumers can exercise their freedom to choose

That only works if the consumer is fully informed.

In the case of bait & switch sometimes the switch happens without them being informed and/or without giving them chance to cancel/uninstall/whatever.

Just a reminder: https://bitworking.org/news/2008/01/the-free-market-fairy/

What about this clause: "The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software"

It sounds like they went out of the way to remove that copyright notice. Am I misinterpreting that part?

Genuinely curious since this stuff affects me....

As the original Copyright holder, you can do that. It just doesn't make the old Apache licensed version disappear.

Only if they do hold all the copyrights. If they've removed the notices from other people's open-source contributions (and don't have a CLA or similar that permits them to do so) then that's copyright infringement.

I thought the MIT license was the original one?

It was Apache2, see this (preserved) fork.

https://github.com/Dramaga11/vsc-material-theme/blob/main/LI...

Changing the historical agreement is fraud.

If he edited the repo history to edit the licence that is.

it's a problem. As soon as it became easy to ask for money via Patreon or githib sponsorship, etc... tons of people are going to try to get some for minimal effort. It's just the nature of the beast.

Asking for money isn’t a problem. The problem is this person went out of their way to extract money by harassing people who rightfully use the open source Apache 2 version, switching the marketplace extension to a closed source version with obfuscated code (likely malicious according to MS), and possibly more, all this for doing a quite small amount of work. That’s after already raising $7.6k, apparently.

Making a theme is not minimal effort which is why 99.9999% of everyone uses preexisting themes rather than making their own.

LoC is a red herring. And the fact that thousands of people decided to introduce his LoC as a dependency in their local development environment makes the point.

I'm surprised to see supposed software engineers get this wrong. We tend to dismiss look&feel type things as low effort which is funny because very few of us could make a make a good theme if we tried, yet it's "just" a matter of picking colors.

Maintaining a theme ~a decade after its creation is minimal effort. You can check the commit history yourself and see it's a one line change once in a while among auto dependency bumps thanks to a >10k line package-lock.json, at least half of which is unrelated to core functionality as demonstrated by t3dotgg.

It's also funny to see a "supposed software engineer" shit on other's tools and post veiled personal attacks based on a single Internet comment. If I were you I'd probably insinuate you've never worked on any large code base so don't know LoC definitely correlates with complexity as a rough first signal (not to mention I included a rough analysis of actual code as well), but I shouldn't do that based on a single comment. Anyway I've maintained plenty of open source code bases large and small, popular (>10k stars, yes that's not a very good metric but it's also a useful first signal) or not, so I'm pretty well equipped to evaluate maintenance burden.

Yeah, pretty much. I am actually baffled. Heck, I am impressed he raised 7k for a theme.

Almost all of the work has already been done by other people for other text editors. Porting a theme requires copying files from another project and writing a tiny bit of glue around it. The author of a niche terminal emulator I'm using on small systems was able to add around a hundred themes this way in a short amount of time. Sadly, he wasn't able to raise $7.6k×100 for all that thankless work.

I think effort is irrelevant. Value is what we really look at when deciding what price to pay. It doesn't matter to most people if it took someone a 1000 hours to produce a loaf of bread. They're not going to pay 100x the price of the bread that took 10 hours to produce. Especially, if the products are mostly indistinguishable.

Partially in jest, but have you forgotten about the luxury market?

I think the luxury market proves my point. The perceived value of luxury items is much higher as reflected in the price paid. Although, the cost and effort to produce luxury items is roughly the same as a similar non luxury item.

hi, maintainer of the fork here

just did a pass and removed everything that was not necessary - it's even less code now lmao

    -------------------------------------------------------------------------------
    Language                     files          blank        comment           code
    -------------------------------------------------------------------------------
    TypeScript                      23             50            169           1307
    Markdown                         6            129              0            224
    YAML                             2              8              6             52
    INI                              1              1              0              7
    -------------------------------------------------------------------------------

    SUM:                            32            188            175           1590
    -------------------------------------------------------------------------------

Reply on news.ycombinator.com