Running a Bakery on Emacs and PostgreSQL (2019) | Hacker News Reader

Running a Bakery on Emacs and PostgreSQL (2019)

bofh.org.uk

193 pointstosh48 comments

I‘ve been using MS Access some twenty years ago to do all sorts of magic for small businesses. That was from the 90s to the early 2000s. - Later I‘d use Python to do magic for companies who didn‘t know anything better than MS Excel to handle large sets of data, and when they were lost in chaos, I‘d start to streamline, process, tidy their data and help them make sense.

Just two examples that showed me: Use whatever helps you to get the job done, automate and streamline processes, select tools that might give you a competitive advantage.

Edit: I love Emacs, for everything text and code and some more, but there are way better tools for the task the author describes.

Emacs (esp. with org-mode) sits close to the barycenter of the space of problems like this, where the vector sum of "effort to use the solution to solve the problem" and "engineering effort needed to build the solution" is quite low. Building a React web app, or a Visual Basic bespoke app, might yield a nicer to use program, but would require more effort than throwing something together with Emacs Lisp, SQLite, and org-mode.

It's one reason why I won't give Emacs up, again despite half of Hackernews being convinced the VSCode will solve world hunger.

I love Emacs and Orgmode, especially in combination with Sqlite. And I am convinced that VS Code - despite doing so many things right - does not have the versatility and the capability of Emacs, no matter how many packages are written for it because it is not intended to be configured by it‘s end users in a way where they read the code of the packages, learn from them, change them according to their needs and automate all kinds of processes / workflows with it. Emacs is an application framework for the user with a hacker mindset. And Emacs Lisp is the underlying scripting languages that invites you to play around with your tool.

Having said that: Yes, you can run your business with Emacs alone. But I am not convinced it is the right tool for the job.

Viscose is impressive, but emacs is a lisp machine.

One problem is you have to learn emacs. Then for employees they have to learn the command line etc... so reduce friction I guess is why we have webapps.

On almost all Emacs posts, I see folks go like,

"<harsh criticism> PS: I love Emacs"

Only vaguely related: what would be the Access equivalent of today? Somethings with low learning curve and forms.

Any suggestions?

Commercial Web-Based Databases: http://quickbase.com https://www.zoho.com/creator/ https://www.knack.com/ https://bubble.io/ https://airtable.com/ https://www.claris.com/filemaker/ https://www.appsheet.com/ https://ninox.com/en https://www.honeycode.aws/ https://www.ragic.com/ https://www.fusioo.com/

Surprisingly, there's not much in the way of open source with the same level of functionality as above. Here's a few:

https://cortezaproject.org/ https://www.openxava.org/ https://www.joget.org/ https://www.nubuilder.com/ https://www.openoffice.org/product/base.html https://baserow.io/

Access still exists, although I haven't used it in a long time

Alternatively, you can access Google Sheets with SQL queries, although I guess it probably wouldn't be great, but Google Forms + Sheets?

Force.com/Salesforce - literally the reason I pushed our business unit to purchase a couple dozen Force.com-only licenses was we had teams that needed really simple apps to store and track data and my team (internal development) did not have the bandwidth to make them with our traditional stacks.

Had something up within 48 hours, still working and worth it to this day.

> what would be the Access equivalent of today

Access.

Sqlite plus some kind of scripting language plus some graphical viewer for that database.

The phpmyadmin(backed by mysql), html4, and apache mod_php platform, probably deserves an mention here as it did play an huge part in the migration from offline PC databases to online webapp's in the space where MS Access was prevalent.

The conventional wisdom in 2021 would probably lean toward python+flask+postgre rather then the traditional lamp stack but i am not sure that's actually an superior solution for the entry level user/designer compared.

We are working on it (starting from the VBA interpreter).

Simple CRUD apps with Python + Flask or Django I‘d say.

Even easier would be to use a CMS, I prefer Processwire but many others also have great form builders and generic list/calendar/etc views.

Airtable

GnuCobol? Text mode only.

Thank you for all the suggestions! Will check them out!

glide

https://www.glideapps.com

Sharepoint forms.

Oracle Apex is probably the best

One of the low- or no-code tools.

Yeah, this is geeky and technically pleasing, but destined to live only as long as the guy has enthusiasm or the will to keep doing this as a small business. As soon as it comes time to have someone take ownership or manage the system, the model breaks. What retail bakery employee is he going to find to maintain the SELECT * FROM statements and look at the Github repo for release notes?

There was a post a few years ago with the message "don't get too clever or special when designing a system if you want it to be maintainable for any reasonable cost by an average programmer". I think this is related...

Edit: the bakery website is gone or down, 2 years after the story.

Meh. It's a system deployed on production that works, and is being used weekly. If he ever gets to the point where he needs to employ someone to do it, the system will be mature, and building a web ui rather straightforward.

Or it might not, but if he's successful to the point of getting employees, he can start worrying about it then. Isn't that like the number one advice for entrepreneurs? Don't worry prematurely about scaling out.

(2019); discussed at the time: https://news.ycombinator.com/item?id=19252952 (191 comments)

I think PostgreSQL is great, but I also think SQLite would be much better for this use case.

What would be even cooler would be a minimally updated FoxPro, just to make it run on modern OSes and such.

xharbour?

This is super cool, but I’d worry a lot about passing on the domain knowledge to somebody else without a dead-simple TUI/GUI. Eventually somebody else will need to run things.

Will this be software the author will ever have to pass on? I'd guess probably not, considering it's software for his small bakery. It's software directly tied to his identity - he's the sole consumers. That's exactly the kind of software emacs is great for, and precisely the kind of software "hacker" types use.

He's not going to sell this as a "bakery management system" and become a software company, he's just going to bake bread. If he needs to share the _information_, then he can just print reports, that's ostensibly what the system is _for_.

If he wants to grow beyond a single shop, he'll either need a standard entry format, or be prepared to compete for employees against enterprises which actually need a high degree of technical literacy.

A software engineer starting a bake shop doesn't exactly sound like someone looking for a "scalable" business.

This reminds me of supermarket clerks running IBM 3270 apps one handed. As long as the UI is straight forward, it doesn’t need to be pretty or slick. It just needs to be second nature.

I'd also worry about bugs accidentally inserting bogus data into it.

The author says he doesn't like spreadsheets because of all the copy and pasting, but I think that's a feature.

You go over every calculation to make sure it's right, and once you have a template sheets to copy and paste for a monthly report or something, you're able to do a lot less of that.

It can be hard to view a raw database to track issues down depending on how it's structured. Usually you need more than viewing tables as spread sheets at least.

Training I guess?

I feel like this is a programmer’s equivalent of using Excel to record UAT results or Configuration decisions. Absolutely possible and is mostly fine, but really isn’t the best tool for the job.

In his follow up post he talks about union combining identical results. If the author ever reads this, you can prevent that by using `union all` instead of just `union`. That won't remove duplicates.

Indeed, though his work-around of adding a path column is something is worth having anyway, because it makes it much easier to see where the generated output has come from.

It is forcing the engine to do more work though, because it demands a distinct result set, so it adds extra nodes to the plan to remove duplicates. If you know your query doesn't have duplicates, it's much smarter to not add an extra whole sort/distinct when totally unnecessary for correct results.

I'm fairly sure that a bakery doesn't produce the amounts of data where you need to worry about such things.

Eh, it's all relative. You can make some queries perform pretty terrible even with tiny data. It's something that has had measurable performance impacts on queries I've had to tune and deal with with over the years.

I don't know, when a baker needs a dozen of something, they tend to bake 13. Maybe things get out of hand.

I love me some Emacs and the aspect of not having one's data rolled up into some inscrutable proprietary binary is certainly appealing.

However, are there viable open-source programs to specifically meet this need (i.e. the daily operations of a small business)? I've seen a lot of startups around this (e.g. ipads at a lot of checkouts now for small businesses), but is there anything that allows the business to not have their data beholden to some 3rd party?

Soon Emacs will replace bread oven.

Well, looks like Emacs is really an operating system disguised as a text editor.

Look ma, no cloud ;)

Why would ever you use anything else.

Reply on news.ycombinator.com