Big Ball of Mud (1999) | Hacker News Reader

Big Ball of Mud (1999)

laputan.org

142 pointssoftwaredoug31 comments

Every time this is posted I forget and think it's going to be about the art form that is, roughly, a big ball of mud that's polished ridiculously smooth.

https://en.wikipedia.org/wiki/Dorodango

...then you remember that most applications are also just one big fragile polished big ball of mud.

One of the best articles about software engineering ever written. They should pin it to the top of the HN front page :P

Agreed. Reading this in 1999 actually explained everything I’d seen and everything I would see. I’ve read a lot of papers but this is the one I see in real life every week for 22 years.

I liked this urbit related blog post that starts with the 'big ball of mud' and thinks about how to avoid this if you were to boil the ocean and redesign everything: https://moronlab.blogspot.com/2010/01/urbit-functional-progr...

This is a great article, but it has some considerable history, on HN: https://news.ycombinator.com/from?site=laputan.org

Reposts are fine after a year or so! https://news.ycombinator.com/newsfaq.html

These seem to be the threads with comments:

Big Ball of Mud (1999) - https://news.ycombinator.com/item?id=28915865 - Oct 2021 (23 comments)

Big Ball of Mud (1999) - https://news.ycombinator.com/item?id=22365496 - Feb 2020 (48 comments)

Big Ball of Mud - https://news.ycombinator.com/item?id=21650011 - Nov 2019 (1 comment)

Big Ball of Mud (1999) - https://news.ycombinator.com/item?id=21484045 - Nov 2019 (1 comment)

Big Ball of Mud (1999) - https://news.ycombinator.com/item?id=13716667 - Feb 2017 (6 comments)

Big Ball of Mud (1999) - https://news.ycombinator.com/item?id=9989424 - Aug 2015 (9 comments)

Big Ball of Mud - https://news.ycombinator.com/item?id=6745991 - Nov 2013 (21 comments)

Big Ball of Mud - https://news.ycombinator.com/item?id=911445 - Oct 2009 (2 comments)

The "Big Ball of Mud" Pattern - https://news.ycombinator.com/item?id=10259 - April 2007 (2 comments)

Thanks for taking the time to do this.

Hey dang, I was wondering. Do you have any kind of automatization for this type of comments? Or do you manually build the repost list on a selected few posts?

Scroll down to the bottom of HN pages, there's a search box; paste the URL in and it shows previous submissions: https://hn.algolia.com/?q=http%3A%2F%2Fwww.laputan.org%2Fmud...

At the top of the page you can click "past" which will do a search on the title, or click the hostname to see all other submissions of that site. But I don't think that's what BaraBatman was asking: rather whether dang has automation for performing those searches and constructing a comment from the results.

I build them manually but can do so pretty quickly using HN Search and a browser extension I wrote that does a lot with keyboard shortcuts.

I may be wrong but isn't the submitter warned that this link has been already posted when creating new submission?

No, reposts are fine after a year or so: https://news.ycombinator.com/newsfaq.html. That's always been HN's policy and it strikes a nice balance - it means the site is not overrun by duplicates, but at the same time good content gets multiple chances at discussion. That's especially important for new cohorts of users, who haven't seen the classics yet.

EWD 215: A Case against the GO TO Statement (Go-to statement considered harmful): https://www.cs.utexas.edu/users/EWD/ewd02xx/EWD215.PDF

The Rise of "Worse is Better"

https://web.stanford.edu/class/cs240/old/sp2014/readings/wor...

The article has its own reference which I'd prefer for its nicer hypertextness and amount of contextual and referenced information:

http://www.laputan.org/gabriel/worse-is-better.html

Seems like a lot of “cloud native” software is this, and nobody suffers but the people who have to work on it or pay the massive cloud bill.

Well, "the only thing that matters in software is the experience of the user", so don't expect much change as long as the UI looks fresh and the users are happy.

> "the only thing that matters in software is the experience of the user"

I think it's true that user experience is all that matters. Software engineering is just the realization that this will also be true 12 months from now.

The authors could not have even imagined the crazy new innovations in mud balls that microservices would create. And much harder to address.

Microservices are the act of getting the your ball empty of mud by taking small pieces, squeezing them as hard as you can, and throwing the results all back at the large ball.

It's like a reverse piñata.

All software eventually becomes this, "cloud native" or otherwise. It is the "everything evolves into crabs" of software.

This describes the app I've been developing for the last couple of years.

I've been working with a nontechnical team, who didn't really know what they wanted, and it "accreted," over a period of about 20 months.

After we had basically decided that we had finalized the features (they finally knew what they wanted), I tossed out the entire old codebase, refactored the business logic into a framework-independent package module, and have been rewriting the UI, with a designer, on top of that.

Every now and then, I can use some snippets from the old codebase, but the majority of the code is new.

It adds about three months to the schedule, but it is totally worth it.

Is the old system still running or do you only keep the codebase around for reference?

It's not a running system. It's an iOS app.

The backend (servers) are running (they haven't changed much), but I keep the old codebase around for snippets.

If we'd have known about Dorodango in 1997, we'd probably have found a way to fit it in...

aka main.cpp

I find the all caps links scattered throughout this article extremely annoying for some reason. Just distracting. Feels very 90s. I'm glad we're 24 years past this style of content.

By contrast, the listicle that you would read today, written by a college student for some mid-tier “technical news” outlet, would eschew any meaningful reflection on the history of software engineering. It would be required to be too short to actually contain a fully realized argument in order to make room for mobile friendly adverts.

Reply on news.ycombinator.com