Hacker News new | past | comments | ask | show | jobs | submit | tiben_'s comments login

One point is code that is highly dependant on underlaying storage system (DBMS etc.) is hard to unit test.


Have the same feel when two months ago i switched from 10 years vim usage to intellij for code. The UI is so well designed, the "everyday's coder problem is so well understood in the UI layer its just a pleasure to work with this tool. The UI just reflect that. Everything is just so well organised. As a web UI developper, and despite material/responsive requirements i refer often to IntelliJ's GUI. Its just so well designed.

OpenSSL can have this kind of UI. Its just a matter of understanding user requirements.


GUI's can be counter productive for theses kind of tool. There are some contexts like "SSL" were the user needs to understand the underlaying concepts a UI cannot teach in itself


Does a config file or a pile of cli flags do a better job of teaching new users the domain?


I believe they do. Configuration files usually have comments, GUIs typically lack that or documentation. If they do not, you have to look it up somewhere. GUIs could have documentation, but it is very rare I would say. It is something similar with the CLI flags as well. That is just my opinion. Maybe I am the only one who reads manual pages that explain what a CLI flag is supposed to do, or for example what key X means in a configuration file.

I have been taught more by those two, than by GUI. The only thing I could imagine that may teach me through GUI is Smalltalk, but that is very limited. On top of that, GUIs might be less common for whatever reasons as well in the UNIX world.


Configuration files occasionally have comments. CLIs sometimes have man pages. GUIs sometimes have tooltips and often have additional help documentation. I'd say that's pretty much a wash. But GUIs can also much more easily convey complex relations between options, such as some option only taking effect when another two each have a specific value, or even simpler things there being a specific set of permitted values for an option. Sure, you can document the cli but that information is probably going to be inherently presented as part of the GUI.


as a mail client connecting to "real life" people i just have to disagree with theses dogmatic statements


+1 cli dogma has limits. I now use Gnome/Evolution which just "works"


* for your usecase, congrats!


I think it answer needs of a lot of people myself included.


I tried and used it 10 years ago in my 20's period were i discoreved the joyful of terminal world. But as of today, no.. it is not sustainable in professional world. As Vim which i used to love for years, but now i must admit IntelliJ simply does the work... mutt, for a mail tool is simply not plug and play. have no time now to deal with configuration files, proxy tools and so on to deal with such a simple thing like sending an email.

Please note i'm not bashing mutt, its a great peace of code


“it is not sustainable in professional world”

Huh? I use email for all my important communication, and Mutt is the only client that is powerful and convenient enough.


With common email providers like gmail, live.com and so on, you have to configure IMAP proxies with is just an headache for this specific use case.

Not to speak about real life use case were the email is HTML formatted with images.

mutt, as far as i know is rooted to Unix local email handling. Which is not very friendly when used with common web providers 99% people use today

In 2022, I love linux and command line. But mutt is outside the limit i can tolerate about this dogma.


I use offlineimap to get my mail to my local machine in maildir format, so no need for any imap etc in mutt. Bonus: I can use tools like cat and grep etc on the emails.

Also, there's a lot of tools to have mutt imap support work with gmail/office365 etc.


cat and grep are not the tools to deal with HTML content.


I think you’re mistaken. Mutt works fine with IMAP servers. It’s quite flexible. In the small minority of cases when I need to see the email in a full-blown web browser, it takes two keystrokes to open it there.


I know this. i've spent a lot of time years ago to configure mutt to fetch my gmail/hotmail/free accounts and it worked pretty well. But when you change computer hardware what a hassle to reconfigure everything. Such a simple thing like sending email should now be a no-brainer. i quit this dogma to use my brain for more insighful things


mutt config is just plain text files. Store them in git/ftp/http server nad grab them to a new machine and everything JustWorks


Imap proxies aren't just a question of conf files.

Email today should be a no-brainer really. Its a hobby to configure mutt.


You dont need imap proxies.

If you want your email locally: Use offlineimap to grab the email for example. There are many ways to connect. If not: there are a lot of python/bash scripts to make imap/whatever work with mutt

Almost all of them use configuration files that can be stored in git/svn/csv/etc


The thing about offlineimap and isync (which I prefer because it is significantly faster in my experience) is they work great until they don't. I originally switched from offlineimap to isync, because of some bug (might have been offlineimap or exchange IMAP support, I don't recall) which caused some email loss. Just this week I had to deal with some isync issue which causes the sync to duplicate mails when uploading to the server. I still haven't figured out what's going on, I'm getting Keywords not supported errors from the server when syncing, which caused this. It happened once before and I had to wipe the whole directory and resync to fix it.

I am now looking at what other solution to use. I used to use muttator which was a plugin for thunderbird that emulated mutt behavior and was great, but that hasn't been maintained for >7 years


offlineimap is what a used. But now i don't have time for the configuration of this. seems today pretty archaic.


I configured mine only on the days I switched jobs. Copy one entry to a new one, changed name, changed password/oauth keys and ran `git add .dotfiles/offlineimap.conf && git commit -m "new work email configs" && git push` and call it a day. This is less time than I spend with ANY other program I use at a new job


How many time you spent the first time you digged into theses tools ? i'm sorry but i'm ready to spend huge amount of time to learn tools like GCC or anything else that's complex. Not for sending mail.


I miss mutt a lot, and would much rather be using it than the weird web messes we have. But I also have to have extremely specific corporate branding in all my outgoing emails. It's stupid as hell but it's also not worth fighting about.


Maybe you could use one of Mutt’s hooks to attach the branding to your outgoing email.


Six hours of wrangling some HTML merging later, I get a message from bizdev@company.com

"Hi morelisp,

In the last email you sent the logo wasn't properly aligned to the left with the text and the font of the phone number was not correct. Please reset your signature block in your Outlook configuration based on the official template."

No thanks.


You actually tried it? Impressed.


No, I'm saying what would inevitably happen.


I once worked in a shop where Exchange was configured not to send text equivalent with HTML emails. Even messages sent in plain text would be munged by Exchange into unreadable HTML hash. These may in fact be the default settings. I emailed the sysadmin asking that they be changed, and his only response was "Try using an email client from this century."

Mutt is NOT suited for the professional world. The professional world has standardized on Outlook and Exchange. If you use something different, no one is going to change anything for your prima donna ass if your emails turn out unreadable. Use Outlook for your company mail.


I have not found any issues using other email clients (including mutt) in our exchange environment. In fact one thing that can be said about outlook is that every single user hates it and their webclient is even worse.


Is the “professional world” a place where you’re paid for your work? Because I’ve been paid for my work for 35 years, and I’ve never touched or dealt with either of those programs. I’ve used Mutt for about 20 of those, and nobody’s complained about my emails, nor have I had any trouble reading others’. As other replies to you have pointed out, your particular experience is not definitive of the standards of some “professional world”. In fact, from your description, it sounds like you’ve been working in some pretty unprofessional places.


> I once worked in a shop where Exchange was configured not to send text equivalent with HTML emails.

You can just plug in an external program to convert HTML to plaintext; IIRC elinks/lynx/w3m can do it.

> "Try using an email client from this century."

Outlook launched in 1997, so obviously that's out.

> Mutt is NOT suited for the professional world. The professional world has standardized on Outlook and Exchange.

Yeah, no; a large chunk of the professional world lives in Microsoft's little bubble, but plenty of companies live outside of it quite happily.


I understand the willignness to do everything in command line. But it's just dogma. Seriously, using so many external tools to deal with such a dumb and common thing like sending/read a mail doesn't have to be so complex. It's useless complexity for nothing. As i recall, mutt has been designed to handle local Unix mail. Not to deal with web providers as the majority of non-nerd people use today. It's a significant effort to configure the tool for this common use case, the soft originally has not been designed for that hense the useless effort to adapt it.

It's not my hobby anymore to spend time around this. I just use the app provided by my telecom provider to send SMS, i haven't configured an obscure TUI tool to interface my local 4G antenna to just tell my wife i'm well. Email is the same. Life is short hence the idea that using mutt is just a nerd hobby today. You can't advocate common mortal to go this way it's just outdated and overall ridiculous.

If its your hobby, well, i'm happy for you


it's not dogma, it's a preference.


Cups has its own set of cli tools, see <https://www.cups.org/doc/admin.html>, and supports "lp" and "lpr" commands, see <https://www.cups.org/doc/options.html>. But theses are not TUI though.


Does anybody knows the artist who made the "slamm" phone poster on the article illustration ? or where can i find it ? I really like it.


The artist is Deborah Azzopardi (https://deborahazzopardi.co.uk/)


Thank you very much


"With rare, old media like that, the best chance to preserving them is reading the raw magnetic flux data with specialized equipment"

Kryoflux is great for that. https://www.kryoflux.com/?page=kf_features


Or if you prefer free-as-in-speech kit, there's also Discferret (www.discferret.org) and a ton of other similar community-developed tools :)

(Full disclosure: I'm one of the guys behind Discferret but it hasn't seen much dev work recently due to lack of time on my part)


Seems the correct project URL is : https://discferret.com

That's interesting i was not aware of this project. I'll take a further look asap. Is there some feature/comparison sheet with Kryoflux available ?


Sorry, quite correct. Apparently I can't even remember my own domain name!

It'll do exactly what a Kryoflux can do, including flux-transition-level read and write. (I'm not sure if the Kryoflux can write). The big differences are that it has a complete Shugart drive interface which matches what floppy drive datasheets advise -- so it's fully open-collector, with the right type of gates.

This also makes it very flexible -- it's been used to read ST506 MFM/RLL drives and decode the resulting output. It's been connected to NEC 8-inch drives with non-Shugart interfaces.

Its only major problem was that it was expensive - about $150 per unit. You could probably get that down a lot by using a more appropriate FPGA. If I did it again, I'd probably use a Lattice MachXO and the Yosys/Nextpnr toolkit, and possibly an STM32 microcontroller.

Practically speaking if you didn't need Winchester support, you could probably use the STM32's timer and PWM peripherals with DMA reload, leaving a single-chip solution with a bit of LSTTL for the drive interface.


Heads up, store.discferret.com isn't resolving.


The store closed years ago. I was working 7am-6pm for a long while and just didn't have the time to put in.

I still have about a dozen boards and a few components, actually. Look me up on Twitter (@philpem) or Mastodon (m0ofx@mastodon.social) if you're interested in a board, my usual offer is "beer money plus shipping" :)


Thank you for theses details. Kryoflux can write back a read raw stream or some common image formats (IFP, ADF, G64 etc.)


Oh, wow. This is pretty cool. I've got some fairly oddball 5 1/4" disks from the 1980's that I'd like to try to preserve, and this looks like it would do the job. I'll echo what another comment says-- the store URL isn't resolving. I'd love to purchase one of these units.

I thought about using Kryoflux in the past, but their licensing is a complete non-starter for me. I'm not going to purchase hardware and license software that limit what I am allowed to do with it.


Unfortunately I don't sell the ready-made units any more -- they're just not cost-effective to sell like that at a price the market will bear.

I want to do a redesign with more cost-effective components (an STM32 Nucleo and a shield or Morpho backpack is one idea I'm considering).

I do have some PCBs left over, a bill of materials and can give you some guidance with part substitutions if you want to DIY one. Feel free to drop me a note on Twitter (@philpem), Mastodon (https://mastodon.social/@m0ofx) or email (via www.philpem.me.uk) if you're interested.


Far better than the original stylesheet (joke)


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: