To be honest I am still too much in a NIP & Kind detail learning phase to give a good opinion on this. We find out, learn by trail & error. Currently it's working basically, but as you suggested it's not a good technical approach we should optimize it. In the future with nostr:nprofile1qy8hwumn8ghj7srwdaejumr0dshszxrhwden5te0ve5kcar9wghxummnw3ezuamfdejj7qgswaehxw309asjumn0wvhxcmmv9uqkvamnwvaz7tmxd9k8getj9ehx7um5wgh8w6twv5hkuur4vgcnqcth0f4ku6n8x4er2mrpdfh8ydfnxsensmnyvduk57tvvachxunwwf68zdt8wfengwf4wc6ryutrxeshwum2xs6hjueh8a38ymmpv33kzum58468yat9qythwumn8ghj7ct5d3shxtnwdaehgu3wd3skuep0qqsdc8dhmzvvclfgf3hwcevla6az3ufu3zq286xe8067s73a56ny4vc770yv8 (and other social bookmarking tools) I would like to see:

- Signing date of the first bookmark per profile (who started bookmarking an URL)

- Signing date of the updated bookmark per profile

- Optimally (speed) filter the social bookmarks on hashtags, bookmark / URL, profiles

- Search within descriptions

- Public & Private bookmarks

- Enhance with tools to optimize social bookmarking such as repeatedly 404 detection removal of bookmark (DVM?), LLM to filter on bookmarks during browsing (browser plugin or so). Idea I had for years , and also saw something similar suggested by nostr:nprofile1qy2hwumn8ghj7un9d3shjtnyv9kh2uewd9hj7qgnwaehxw309ac82unsd3jhqct89ejhxtcpr9mhxue69uhkwun0w4c8xtnxd9shg6npvchxxmmd9uq3uamnwvaz7tmwdaehgu3dwp6kytnhv4kxcmmjv3jhytnwv46z7qgkwaehxw309aex2mrp0yhx6mmnw3ezuur4vghsqgrkcud2uwjfruweamz8ewshug5umfq38g9mkmn2u9mk6ajru2w2lg26q8zu