AskHN Digest

1. Ask HN: What is the best method for turning a scanned book as a PDF into text?

Top comment by aragonite

I did this very recently for a 19th century book in German with occasionally some Greek. The method that produces the highest level of accuracy I've found is to use ImageMagick to extract each page as a image, then send each image file to Claude Sonnet (encoded as base64) with a simple user prompt like "Transcribe the complete text from this image verbatim with no additional commentary or explanations". The whole thing is completed in under an hour & the result is near perfect and certainly much better than from standard OCR softwares.

2. Ask HN: Is anyone still using Dreamweaver?

Top comment by neom

I do!!!!!! I love dreamwaver even today, I'm surprised people don't use it, they have done an amazing job keeping it up to date - it's honestly a joy to use. Granted: I'm not a real dev/swe, just a dude who likes to mess around with webtech, still, I think "real devs" would enjoy it too, it's great to use.

I learn web on dreamweaver, I would make something on the front end WYSIWYG editor, and then "turn it around" (I called it in my head) and look at the "back of it" (I was a kid) - anyway, tables and frames and dhtml baby!!!!

Also: https://s.h4x.club/nOu445qL :) :)

3. Ask HN: Tired of startups – want a normal job. Help

Top comment by bdcravens

> I feel like a freaking loser given my age and my relative career stagnation / my inability to actually turn my mildly above average skills into any real money.

There's a lot of noise out there that tells us we're supposed to get rich off of entrepreneurship or VC money, and not enough telling of the truth that those success stories are pretty rare. No one who is doing something they at least minimally enjoy and not breaking their body doing it, and getting paid in the top 20% of income earners, is a loser.

4. Ask HN: Small Ideas vs. Big Ideas?

Top comment by speakfreely

There's a "Growing B2B SaaS" podcast that had a guest that made an impression on me. He said tech is the wrong industry to try to grow a lifestyle business. You can do it, but your business is substantially more likely to be disrupted than more brick and mortar industries.

You might've had a solid calendar scheduling app in 2015, but by 2024 everyone is eating your lunch from every angle, while the plumbers continue to plumb. And the major moats like network effects or marketing specialization usually only apply to larger businesses.

His advice was to always go big, grow quickly, and sell quickly. Even if you go smaller for a higher chance success, the income stream likely will crumble as soon as you stop actively pressing forward with the product because of the intense competition in the software space. The lower risk option of growing and selling quickly, in his estimation, was the best risk-adjusted return.

5. Ask HN: Can I really create a company around my open-source software?

Top comment by bruce511

I get that you've worked on this for months, that you're burned out generally, and now unemployed. So this comment is not meant as "mean" but rather offered in the spirit of encouragement.

Firstly, building a business (especially in a crowded space) is stressful. It's not a place to recover from burnout. It's not a place that reduces anxiety. So my first recommendation is to relax a bit, put this on the back burner, and when you're ready go look for your next job.

Secondly, treat this project as an education. You had an idea and spent months implementing it. That's the easy part. The hard part is finding a market willing to pay money for something.

So for your next project do the hard part first. First find a market, find out what they will spend, ideally collect a small deposit (to prove they're serious) and then go from there.

In my business we have 3 main product lines. The first 2 happened because the market paid us to build a solution. We iterated on those for 30 years, and we now are big players (in very niche spaces.)

The 3rd happened as a take-over of a project by another retiring developer. He had a few customers, and a good product, but in a crowded space where there's lots of reasons not to change. It's taken many years to build it out, despite being clearly better than the competition, and it's still barely profitable (if you ignore a bunch of expenses paid by the whole business. )

The lesson being to follow the money, not the idea. (Aside, early on we followed some ideas, all those projects died, most without generating any revenue.)

So congratulations to seeing something through to release. But turning a product into a business is really hard. Turning a commodity like this into a business is almost impossible.

I wish you well in your future endeavors.

6. Ask HN: Why isn't anyone else bothered by the fact that life is only on Earth?

Top comment by skissane

> I am bothered by how improbable it is that even simple life forms have not been discovered anywhere else in the universe.

The obvious reason is that our ability to look is extremely limited.

Could there be subterranean life surviving deep in Martian rock? We don't know, we haven't looked yet there yet.

Could there be life sustained by thermal vents in Europa's subterranean ocean? We don't know, we haven't looked there yet.

We've identified some habitable zone exoplanets. Do any of them have life? We don't know, we don't have the technology to tell. If we are talking about Earth-like multicellular life, you'd expect there would be some atmospheric changes, which we'd eventually be able to detect remotely via space telescopes (even if the current generation of space telescopes can't yet). But if we are talking about subterranean microscopic life – how could we tell without physically sending a probe there? So I don't think we'll know this century.

And surely there's many more habitable zone exoplanets out there we haven't identified yet, because our telescopes aren't good enough. And eventually we'll get better telescopes, and we'll identify more of them. But even then there will be many more that even those telescopes aren't powerful enough to identify.

7. Ask HN: Can we get a HN background color (dark mode)

Top comment by roskelld

I use my own CSS to override the default look.

I added thumbs up emoji for voting, book emoji for read comments, larger user names, larger headers, and dark color scheme.

https://postimg.cc/f3nmR6DG

https://pastebin.com/zeXGi4WQ

8. Ask HN: Do you still use Google?

Top comment by sunday_serif

I use google search only for the most trivial searches now. Something where I know almost precisely what information I want.

I use LLMs for most info now, any slightly ambiguous query really. Often times I use the LLM to figure out what source I should find, then I just use google for retrieval

When I want to “surf the web” I use kagi search (usually with small web filter) when I want to see sources written by people. This isn’t that often, but when I need it kagi is the best!

An example, the other day I wanted to make fermented hot sauce, and I couldn’t remember how much salt was needed for the fermentation. I could google and get served hundreds of crappy ad ridden recipe sites that have the answer to my question buried under five pop up videos, or I could ask an llm with one or two follow ups and have a much more pleasant experience.

Then when I decided I wanted to follow a real recipe I used kagi with small web to find a recipe page that wasn’t an ad farm.

A few more steps than google, but all and all so much better.

9. Ask HN: Ways to progress career wise as SWE?

Top comment by Apreche

The easiest way to move up is to change jobs. The biggest mistake I made was staying in one place too long. Then I changed employers a couple times. Got big upgrades in compensation and negotiated a four day work week.

If you don’t climb the ladder at your current job, and you don’t see co-workers climbing, that means climbing is not possible at that employer. Do not stay there more than 2–3 years.

I was afraid to change jobs because I felt safe and secure in my current position. I thought that going elsewhere would be a horror show. My friends often complained, truthfully, about how terrible their jobs were. Mine was way better than that, so I stayed put. But then when I eventually got the courage to change, every time I moved it got better and better.

10. Ask HN: What commit message conventions do you follow?

Top comment by FrenchyJiby

I've written mine up as I've noticed I was repeating myself a bit once my opinions stabilized.

See https://jiby.tech/post/my-git-worfklow/