Comment by Birch-san | Hacker News Reader

Comment by Birch-san | Hacker News Reader

I'm concerned by the team-shaming [0]:

> The only issue left is that of preventable costs arising from error, oversight, or incompetence

> [notably,] the indiscriminate destruction of data […] under the mistaken impression that it was otiose, by a programmer whose experience was insufficient for the task at hand.

> Although he might have been the 'cause of more harm than good' (in the opinion of another of our programmers), […] and while correction of the mistake took some time and effort, it did not result in schedule slippage.

His assertion that "the only phone suitable for the task (of talking to 3 people)" is a crowdfunded phone that doesn't exist yet, shows (at best) poor risk-management.

I don't understand how one developer could need $170/month of mobile data to fulfil their commitments to this project.

I don't understand why the Enterprise edition of Visual Studio would be necessary… nor why Visual Studio would be the right choice for macOS development.

I don't understand why he's being so open with such indefensible spending.

The project seems out-of-touch. One of his roles is to inspire collaboration, so he would have more luck using technologies that excite people today.

[0] https://www.kickstarter.com/projects/1313324524/hermes-eudor...

You only have to see how he writes. What other projects / teams would have reported as "fixed some bugs and replaced the unsupported spell checker technology with hunspell, all going well!" here it takes several pages of... fluff, if you ask me. Either that, or contrived phrases just for the sake of writing a lot of latin expressions.

You never get too much latin in your product updates!!

I had to look up "otiose". Not a good sign for an update to backers.

Fixed some bugs? The project didn't compile under a modern compiler. The compiled version didn't work well with any language besides English or Latin. There was no spell-checker included with the source and we had to re-write the library. Then there was the total inability to negotiate an SSL/TLS connexion... don't get me started.

I wrote for a general audience. Yes, it's a bit fluffy, but I had to try to explain certain things to people who don't know the first thing about computers (if you don't believe me, take a look at the eudora listmoms listserv). The responses I received were overwhelmingly "don't change a thing but the Unicode handling and the SSL!"

> poor risk-management

This crowd-funding itself was an exercise in poor risk management. I mean, you don't need a detailed feasibility study to tell you that starting from a couple ancient codebases based on (1) M$ MFC and (2) Apple Carbon, and ending up with a high-performing, multi-platform application that people would actually want to use today is going to be very hard and probably entirely unrealistic. A cursory look at the code should suffice.

Note that this is the Kickstarter for the Windows version of HERMES Mail. It's an entirely different codebase and, as such, Visual Studio is the perfect IDE for the job.

The high mobile data cost was because said developer lives in the arse end of the middle of nowhere, and he refuses to use Mercurial properly (does a full hg clone instead of hg pull every time he contributes, drives me bonkers to be honest but the man writes good code!)

Said developer also admitted, publicly, to using an unlicenced copy of VS Enterprise after a disk crash. That was an indefensible state of affairs to me, and so the only thing I could do without wasting scads of time was to buy him a licence.

Please don't conflate HERMES Mail with Mail/X. They have similar user interfaces and are mail clients formerly written by different developers at the same company, but they are in different languages, different devs as stated before, and target different platforms.

The finances do seem very strange. Spending $850 on the Cosmo Communicator phone is definitely not necessary, multiple expensive VS Enterprise licenses probably aren’t necessary, and $170 of mobile data/month for one person seems very excessive, at least for the prices in the part of the world that I live in (not unreasonable for four unlimited lines, but far too much for one person).

>The project seems out-of-touch. One of his roles is to inspire collaboration, so he would have more luck using technologies that excite people today.

I hope you're not saying that Eudora should be written using Electron...

:) depends on the desired result.

If the primary threat to the project was "not finishing", then I'd absolutely use Electron. Development with web technologies is fast.

But if performance _is_ critical, and the biggest risk to the project is "lack of community interest", then I'd go with React Native†.

† I'm prepared for the possibility that some newer or more exciting technology exists by now.

Ah, your comments are in jest. Sorry.

Reply on news.ycombinator.com