Where to Buy Hardware

I react to spending money like some people react to being stung by bees. I get inflamed. Recovery takes a while.

If you want to buy a computer or parts, and if you know what you want, so you don’t need a sales representative to “advise” you, here’s what’s worked for me, most of the time (I’m in the U.S.):

  • Newegg.com (an online retailer)

  • Amazon.com (an online retailer)

  • Froogle.com or, if you put on formal clothing to read this page, then the same thing as shopping.Google.com (the newer name and it searches a large variety of online retailers and links to their offers (Google itself does not do the selling))

In years past, I used the Usenet and Web searches to find someone local who’s selling something, but, if you do that, plan on visiting and bringing a bootable live operating system CD to test it (with the seller’s permission but don’t buy without testing unless it’s extremely cheap and still looks plausibly worth the chance). I haven’t tried that in a few years and markets may have evolved. If I were trying it now, I’d probably look in Craigslist, too. People selling computers through the Usenet (not through Craigslist) are probably knowledgeable geeks, since the Usenet is a less-well-known specialized part of the Internet, older than the Web.

When I search, I sort the results by price, lowest first. This gives some peculiar results, because a few bizarre offers use the same description keywords as what you really want, but it’s still the best sortation. You can set a minimum price before searching if you’re pretty sure you won't find a working moon rocket for $5, but I tend to be hopeful, so I start at $0 or blank.

I often ignore other criteria for searches, preferring to rely on my search terms. For example, I usually don’t specify sellers, ratings, or that I must have deals. I look at those things on the product page. If the price is attractive, I don’t care if it’s this week’s deal.

I also don’t specify variables like RAM (memory) size. For some reason, ranges are not clearly indicated, so I haven’t tried those kinds of criteria. What’s on the product page is good enough. And, if the product page is silent about something important, that’s a red flag, especially when other product pages have the needed information. If someone selling omits an important fact, that’s a silent warning.

While ratings don’t much matter, comments matter more. But used or refurbished products are themselves inconsistent, even for one model of computer, so reviews may not be talking about the same thing even when they seem to be. And they could come from people who lack the computer skills or expectations you have, so, while they may be frustrated and angry, they may be making criticisms that you wouldn’t have made because you understand the subject. They may be sincere, but still unreliable. Nonetheless, check them for facts to watch out for, and decide which factual claims are believable.

Many customers want only highly-rated products. As far as I’m concerned, that’s fine for ratings from Consumer Reports or another source that has good criteria for their ratings, but most of the online rating systems have little for criteria, so usually I don’t count the stars.

Then, there are some reviews written just to be for or against the company selling the thing. The reviews are supposedly about the product but are really for or against the company. The reviewers are just promoters regardless of quality or haters regardless of quality. Many of them have been paid or rewarded by a seller, like with a future discount. Their comments are hardly reliable at all, but they’re mixed in with the legitimate reviews and we have to weed them out.

Try to figure out which reviews are secretly from the company’s own flaks or their competitors. One study estimated that fake reviews are around three out of five reviews for certain kinds of products, but, for electronics, those seem to be lower-price product categories, like earbuds. My guess is that real customers don’t bother posting reviews of products that cost only two to twenty dollars. One website analyzes reviews for you, but it doesn’t say how, so I don’t know their criteria. I entered the URLs for a couple of laptops in the $200 range and with thousands or hundreds of reviews at Amazon and that analysis claimed that most reviews were legit. Two product URLs are too small a sample to be a scientific test, but the result is consistent with my view that fakery is less of a problem with products that are not so inexpensive, again probably because people spending significant money are more likely to tell the rest of us what they really think by posting reviews of their own, outnumbering the fakes.

The fake reviews (shilling or hostile) likely will be those with the most extreme ratings and which have only short comments that say little (like “great/lousy product do/don’t buy”). For example, if reviews can have anywhere from 1 to 5 stars, pay less attention to 1-star and 5-star reviews, but look mainly at reviews that have 2, 3, or 4 stars. Look especially at those that say specific things about features, especially combining pro and con. The useless reviews are probably paid for with a requirement to write them cheaply, quickly, and in mass, which is why they’ll be short, hardly informative, and with extreme ratings. Look at the others.

Way-low-price items I typically buy locally, in brick-and-mortar stores. Ordering from an online supplier that has no physical presence in the state where I live can still cost me sales tax, plus a shipping and handling charge applies, so it can be worth paying the sales tax, saving the shipping, and walking out the door with my purchase in hand. Compare buying both ways for your purchases.

Linux-compatible Wi-Fi adapters or network interface cards (NICs) for USB are in a separate article.