Harun

Harun

@harun-sevinc.com

Software Developer - Not a Java-Champion Java, Software-Architecture and Software-Development

320 followers550 following50 posts

Harun reposted

Die Bundesregierung will das Informationsfreiheitsgesetz faktisch abschaffen – doch für ihre Begründung hat sie nicht einen einzigen Beleg. Das musste das Bundesinnenministerium nun vor Gericht zugeben, nachdem wir auf Auskunft geklagt hatten. fragdenstaat.de/artikel/klag...

fragdenstaat.deAngriff aufs IFG: Keine Belege, nix, nirgendsDie Bundesregierung begründet ihren Plan zur Abschaffung des Informationsfreiheitsgesetzes mit angeblichen Sicherheitsbedrohungen. Das Innenministerium wehrte sich bis vor Gericht dagegen, Belege dafü...

„Nach Angaben von Lobbycontrol gibt die US-Tech-Industrie inzwischen jährlich rund 150 Millionen Euro für Lobbyarbeit in Brüssel aus. Sie beschäftigt fast 900 Vollzeit-Lobbyisten – das sind mehr, als es Europaabgeordnete gibt.“ www.deutschlandfunk.de/open-source-...

deutschlandfunk.deSo werden Deutschlands Behörden mit Open Source unabhängiger von MicrosoftSoftware aus den USA, Computer-Chips aus China. Deutschland ist technisch von anderen Ländern abhängig. Das soll sich ändern, auch mit Open-Source-Software.

Harun reposted

Nice way to end the year #Java This flaw is pretty deep in the software stack -- you may not have heard of Mina (few use it directly) but the product you're using might have a dependency to it. Patch away. #software #security thehackernews.com/2024/12/apac...

thehackernews.comApache MINA CVE-2024-52046: CVSS 10.0 Flaw Enables RCE via Unsafe SerializationCritical Apache MINA flaw CVE-2024-52046 with CVSS 10.0 enables RCE via serialization flaws. Patch required.

On Friday evening, we received an email from a user asking if we offer CSV files to download the extracted data from our #webscraper. The answer was, “Not yet, but…”

This post is unavailable.

Some of these tips where used to build our #scraping platform. It took some time but was definitely worth it.

This post is unavailable.

I wanted to share my experience with our #ProductHunt launch. On Monday, we relaunched Adcolabs-Scraper. Technically, it was a relaunch—we originally launched in February but released a major update a few days ago that revamped the entire application. Here are some facts and lessons learned:

View full thread

And, of course, the classic email about a “security vulnerability” discovery, asking if we have a bounty program. Overall, nothing too surprising, but still a bit underwhelming. However, I think we’ll benefit from the backlinks—we’ll see how they’re valued.

On the bright side, we gained 30 new users, which is encouraging! Next step: getting early feedback from these users to see what they liked (or didn’t). 🚀

I wanted to share my experience with our #ProductHunt launch. On Monday, we relaunched Adcolabs-Scraper. Technically, it was a relaunch—we originally launched in February but released a major update a few days ago that revamped the entire application. Here are some facts and lessons learned:

- We finished at #10 this time (compared to #6 last time). This time, we didn’t run any ads on LinkedIn or Reddit, so I’m relatively happy with the result. - We had over 250 unique visitors on launch day, around 240 the day after, and a total of over 500 visitors so far.

- Around 10 people subscribed to our newsletter. - I gained many new followers on LinkedIn. - I received tons of LinkedIn messages offering to boost our launch. - Also received plenty of pitches for marketing, recruitment, UX design, and software development services.

Here we go! We are releasing our new major version at #Producthunt. We made a lot of changes and optimized many things from UI to API level.

This post is unavailable.

Just finished my latest blog post about #Privacy and #Browser #Fingerprinting. Learn why #fingerprinting is a major privacy challenge, what makes it hard to fight, and the best tools & tricks to stay anonymous online. #PrivacyMatters #CyberSecurity #BrowserFingerprinting #OnlinePrivacy #WebTracking

Proud to announce the relaunch of Adcolabs-Scraper in just one week! 🎉 Built with passion to make #WebScraping accessible for everyone. Try our #API or #WebApp and create #Scrapers in seconds. 🚀 Follow along for updates! 🙌

This post is unavailable.

Five days ago, I started a tutorial on YouTube and only watched it for a few minutes. The tab is still open and untouched since then. It follows me everywhere, and now I don’t have the courage to close it.

Just surpassed my all-time record follower count from Twitter! It might not be much, but for me, it's a big deal since I spent years there, and gaining so many followers in just 48 hours feels wild. Thank you, everyone!🥳🎉🙏

Wow, thanks for all the new followers! You guys are awesome—I really appreciate the support! 🎉 Let me take a moment to introduce myself: 👋 My name is Harun, and I’m a #developer based in Germany with nearly a decade of experience, mainly working with #Java and #TypeScript.

I stepped away from Twitter in 2020, but now I’m back here at #bluesky and loving the connections with real, passionate people here. ❤️ I’ll be sharing what I’ve learned, and I’d love to connect! Have any questions? Feel free to DM me anytime. Let’s grow and learn together! 💬

Really excited since we worked the last 6 months on this project beside our 9-5. A lot of work has gone into this update and I hope it gets some love.

This post is unavailable.

1/5 For several years, we have dedicated significant time and effort to #webscraping and its various use cases. Along the way, we've gained valuable insights, and I’d like to share some of them with you:

View full thread

4/5 - Infrastructure is key when it comes to costs. While you can get free IPs from the cloud provider you’re using, they often come with the risk of poor reputation. Don’t underestimate runtime costs in some cloud services.

5/5 Success in web scraping comes down to balancing efficiency, cost, and reliability. By optimizing your approach and infrastructure, you can achieve great results without unnecessary overhead. Happy scraping! 🚀 #webscraping #tech

1/5 For several years, we have dedicated significant time and effort to #webscraping and its various use cases. Along the way, we've gained valuable insights, and I’d like to share some of them with you:

2/5 - Before scraping a page, check if an API is available. - Avoid overloading the page with requests; you don’t need a separate request for every value. - For complex pages with lazy loading or heavy JavaScript, consider using browser automation.

3/5 - Some IPs are better than others: datacenter IPs may suffice in some cases, so use only what you need. - The better your fingerprint (a complex topic), the higher your success rate.

Everytime when I try out some new framework or technology: 1. What is this? 2. Oh, I get it! 3. Wait, no I don’t. 4. Ah, it works! 5. Why does it work? 6. Why doesn’t it work anymore? Repeat forever.

You are up to date.