Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

AI can explain most topics better than I can

This one is demonstrably false. Your personal written style is what's important. Also, you have hands-on experience, which is also demonstrably more than any "AI" has. I urge you to ignore this kind of doubt or consideration.

- What kinds of posts actually worked (for learning, career, network, opportunities)?

For learning, "book report" type posts, just to solidify what I've read in my mind, maybe drive a little experimentation to ensure I've concluded correctly. I've decided not to collect any metrics so that I don't follow from behind, so that I don't end up doing clickbait. Career and network opportunities have not arisen from my blog.

My "public notebook" posts get more traffic, and I've referred back to them, but for me, these are mostly Linux sysadmin topics. I'd wager these are most valuable to people that find them for very specific problems, like seeing LLDP info from inside a WiFi access point or fixing GRUB problems on particular hardware.

- Any practical format that lowers the bar (length, cadence, themes)?

I have not discovered anything for this, alas. I use Hugo, I have a couple of little shell scripts to do monthly counts of finished vs draft articles. I try to stay at or above 5 posts a month. I'm not sure that helps lower the bar, which I interpret as "provide motivation to post".

What would I do differently? Start a blog years before I actually did so.

I'm happy to correspond, my email is in my HN profile.





I second your comment about referring back to sysadmin posts. I do this all the time! Sometimes I even find my own old blogposts in Google.

And I still get a steady trickle of grateful comments/emails in response to a tossed-off post about getting Linux scanner drivers working, many of which are genuinely moving to read.


Hey, just wanted to let you know that your honeypot data for PHP-based attacks ended up factoring in charges being dropped in two criminal cases where prosecution attempted to run with a harebrained and ludicrous theory that basically centered around some... mythical idea of how these attacks happen and how specific they can be. The criminal justice system is where lurid fantasies of how tech works end up putting people in prison for sometimes years and budget concerns meant that attorneys filled parking meters every 4 hours and we had two full time investigators in an office of 40, most with a 80-120 caseload (rolling basis). Sometimes the data one puts online can really make an impact that my guess is that it was entirely unexpected and for two people in their 20s with young kids (separate cases, in fact, although not too far apart), it really reclaimed a good chunk of their lives. Thank you for that, and I hope others would do the same, because one never knows when it'll come in handy. So many products sold to LE are basically snake oil and without data and facts, the threat is incredibly coercive. Any leverage for defense helps balancing the playing field and frankly, nobody deserves to be taken to trial based on utter BS that has merit merely because it matches the equally unfounded anxieties of people, however unsubstantiated.Thanks again!



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

Search: