TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

I Beat the Berlin Rental Market with a Python Script

67 pointsby giansegatoabout 4 years ago

11 comments

silexiaabout 4 years ago
This is why regulation of housing introduces more nightmares than it solves. Rent control eliminates used inventory. Building regulations limit new inventory.<p>Solve it by changing the laws to allow new buildings of all sorts, especially very small and cheap apartments. And by getting rid of rent control.
评论 #26657136 未加载
Jermaine_Jabiabout 4 years ago
I did something similar with Kijiji in Montreal to find a decently priced apartment. It ended up working quite nicely and I found a 3 1&#x2F;2 lease transfer well located in an older building for 700$. I&#x27;ve since employed it for other items that I kind of want, but not enough to trawl through Kijiji regularly for.<p>For reference it was easy to set up because Kijiji thankfully still has RSS and getting timely notifications through Huginn <a href="https:&#x2F;&#x2F;github.com&#x2F;huginn&#x2F;huginn" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;huginn&#x2F;huginn</a> and Pushover was simple.<p>Facebook classifieds seems to ve eating the market a bit, I haven&#x27;t tried scraping fb, I imagine there are barriers, is this true?
gbronnerabout 4 years ago
So the relatively affordable apartments go to those with the best tech skills... Gotta love rent controls. This is exactly like SF in 2000
评论 #26657952 未加载
评论 #26673116 未加载
peter_d_shermanabout 4 years ago
&gt;&quot;Berlin has a massive undersupply of housing — a widespread phenomenon in western cities.<p>Usually, this means that prices tend to increase. That’s what used to happen in Berlin too. Then, a couple of years ago, the city municipality passed a controversial rent control policy that prevents price increase for at least 5 years.<p>It was a very bad idea: the market completely froze and became illiquid.<p>Housing supply contracted (instead of expanding), due to lack of financial incentives.&quot;<p>PDS: <i>I&#x27;ve seen this exact same effect before, in other housing markets...</i><p>Also, a psychological observation:<p>Once people who have had a housing problem in the past find an adequate housing situation for themselves, they, not seeing housing as a problem anymore, are no longer inclined to vote for the politicians and&#x2F;or public housing policies that would help large amounts of <i>other</i> people with their housing problems...<p>Housing, that is, getting the issues around housing exactly correct from a law&#x2F;government&#x2F;macroeconomic&#x2F;public policy perspective -- with a rapidly expanding world population and many homeless people in the world&#x27;s cities -- is, or will be, I think, as future historians look back in time:<p><i>One of the defining issues of Earth&#x27;s 21st Century History...</i>
turtlebitsabout 4 years ago
I&#x27;ve been looking for land for a while, and often thought of making a tool will let you input the normal filters for property search, but include things like distance&#x2F;time to specific locations (Library, Grocery, Home Depot, Costco, etc.), as well as to major freeways. Other ideas including giving ratings based on terrain, flood plains, etc.
mgerullisabout 4 years ago
&gt; I honestly have no clue if there’s a market for this product.<p>I bet you there is. Try offering it as a service of some sort.
评论 #26652163 未加载
评论 #26650287 未加载
Eloraabout 4 years ago
This is a really neat project! I can follow and dupe everything in the writeup until the OP casually mentions:<p>&gt; To get these numbers I trained two supervised models. I fine-tuned the pipeline and hyperparameters with genetic programming on an AWS GPU-heavy machine.<p>Can anyone recommend some ways to level up a bit in this domain to feel more comfortable? I&#x27;m somewhat familiar and understand the underlying concepts to a certain extent, such as supervised vs nonsupervised, but need hands-on practice, what do you recommend?
margorabout 4 years ago
I&#x27;ve done something similar (not in German market) but instead decided to have a chat bot inform me whenever there was a new offer matching my criterias and it also helped me find a flat cheap and in the place I wanted. But your post inspires me to improve that further :)
ivoecpereiraabout 4 years ago
I have done some scraping on my own of some real estate listing websites on my city and applied some filters.<p>The OP goes much deeper into that idea and I can totally see the need for such a thing. I didn&#x27;t go much deeper in the idea as I have found something something interesting meanwhile, but this is a need, for sure.
captainmuonabout 4 years ago
Really cool use of data. I wonder how you want to make an app out of it though without immediately getting legal trouble (Abmahnung) since you scraped the house listings. A friend of mine did something similar for research purposes and there it was a huge grey zone.
bogomipzabout 4 years ago
The post states: &gt;&quot;There are two elements that influence your ability to be selected and get the contract: your profile, and your speed.&quot;<p>Can someone say what &quot;profile&quot; the author is referring to here?<p>Also the apartment photo in the post shows a newish looking almost 64 square meter place with a balcony in Friedrichshain-Kreuzberg on Karl Marx Allee for 1497.56 Euros. This seems like a decent location in a good neighborhood with a decent amount of space. Is this considered expensive for Berlin right now?
评论 #26649349 未加载
评论 #26649072 未加载
评论 #26667337 未加载
评论 #26648285 未加载