Michael Vito・マイケル ヴィート
banner
michaelvito.bsky.social
Michael Vito・マイケル ヴィート
@michaelvito.bsky.social
Place-based engagement, transit, urbanism, anime pilgrimage (seichijunrei, butaitanbou), anime tourism, ramen. 街づくり、公共交通機関、アーバニズム、聖地巡礼、舞台探訪、コンテンツツーリズ、ラーメン。

Shanghai・EN/日本語/中文 https://likeafishinwater.com/
AI scrapers are in general much more aggressive that search engine spiders, and Bytespider has become notorious for being the worst of the bunch by a wide margin.
November 11, 2025 at 7:29 AM
The use without consent is bad enough, but in this case the scraping is so hyper-aggressive it slows sites or takes them offline, just like a DDoS, and creates huge costs for sites that pay based on data transfer volume.
November 11, 2025 at 7:26 AM
Scrape everything. But they really seem to like images (the stats on my CDN were crazy). Based on what I'm reading from other smallish sites' experiences, AI really likes large repositories, so images, code, forums, etc.
November 11, 2025 at 6:48 AM
I have had to put aside time to fight off their scrapers every day for the past three years. drewdevault.com/2025/03/17/2...
Please stop externalizing your costs directly into my face
drewdevault.com
November 10, 2025 at 12:40 AM
This is my feeling about anime to live-action remakes in general. It either loses its magic and becomes a mechanical exercise (JP, usually)... or it's a garish "cartoon, but in real life" (Hollywood, usually). Each is disappointing in its own way.
November 9, 2025 at 11:39 PM
Reposted by Michael Vito・マイケル ヴィート
This time I also had the opportunity to pick up some snacks for the flight home at Chez Inoue and visit the second Nagato Tobidashi at Cafe YONE.
November 8, 2025 at 7:52 AM
I also looked into Cloudflare. They are at early stage still. You can block the scrapers that self-identify, and they are just starting to block those that spoof UA with machine learning, but they still miss a lot of things like this.
November 6, 2025 at 10:47 AM
Can't rate limit. Distributed IP addresses.
November 6, 2025 at 10:45 AM
You can get around the temporary blocks by changing to different browsers, different OS/platform, or using a browser plugin to spoof your own user agent.
November 6, 2025 at 10:34 AM
As a temporary measure, I've taken the nuclear option of blocking the specific user agents, which include recent versions of Safari and Chrome on Mac OS, Chrome and Edge on Windows 10. Other versions of those browsers, or other platforms (mobile, for example), are unaffected.
November 6, 2025 at 10:34 AM
If it's lively now that's good. Was kind of desolate back in 2013. If you were walking around with a camera you would definitely stand out and be noticed (^^;)
November 6, 2025 at 6:04 AM
You took it inside and it is now your kitten? (^^)
November 6, 2025 at 5:56 AM
As time has passed and the boom subsided, the norms with regard to this location have relaxed a bit. But at the time of broadcast, people deliberately did not share the name of or directions to this location, because it is integrated with residential apartments.
November 6, 2025 at 5:50 AM