from Hacker News

Searching for deals on tech stuff isn't so fun. Some sites are heavy on SEO chum and posting as many keyword-laden headlines as possible. Aggregator sites are flooded with whatever products the retailer APIs and feeds push out. I got fed up with all this while shopping online last weekend, so set about making what I'd hoped to find in the first place: a basic, somewhat-automated list of deals

It's perfectly boring HTML with automatic price checks. There's no ads, no cookies, no signups, no writing for machines. The idea is to strike a balance between completeness and simplicity, so people can pick a decent deal without ado whenever they need one. The latest prices are fetched from Amazon, and links to the Wayback Machine are included so users can check for deceptive "yo-yo" deals.

Another minimalist project of mine, txt.fyi, benefited hugely from feedback here way back when.

I suppose the obvious shortcomings are:

• further refinement might inevitably drag it toward more content (risk competing with news media) or more automation (risk becoming a junk aggregator)

• It's limited to stuff in my interests (gadgets!) that I feel confident making snap judgments on whether the brand is good and the deal is worth it.

• Reviewing all the deals every day could become gruelling work. Detecting good deals automatically really seems called for, but ingesting inappropriate or low-quality ones would only create the same problems of trust and frustration that inspired the project in the first place.