←back to thread

270 points ilamont | 6 comments | | HN request time: 1.326s | source | bottom
Show context
jasonpbecker ◴[] No.21973446[source]
Goodreads is desperately in need of a strong competitor.
replies(4): >>21973472 #>>21973686 #>>21974677 #>>21975169 #
bryanrasmussen ◴[] No.21973472[source]
Aside from the spoofing issues what are the main drawbacks and benefits of GoodReads from your perspective - what's the worst of times, what's the best of times?
replies(6): >>21973507 #>>21973634 #>>21973710 #>>21973915 #>>21973926 #>>21974301 #
1. yazan94 ◴[] No.21973507[source]
I want to second this question - this is a relatively easy-to-fix issue on GR's side to have more moderation tools and powers. But otherwise, are there any other substantial complaints?
replies(2): >>21973652 #>>21974724 #
2. zem ◴[] No.21973652[source]
their search and filtering mechanisms are abysmal. they have essentially taken tons of valuable user-supplied data and locked it up behind a useless interface. some simple examples - i cannot find humorous fantasy books by searching for books tagged both "fantasy" and "humour". i cannot do a search that returns hundreds of results and then sort them so that the best ones go to the top - and i'm not even talking about some magic relevance algorithm, just sorting by explicit data. i cannot even organise my own read, unread and to-read books easily. all in all it's a usability disaster and totally wastes the labour people have put into building up the database.
replies(1): >>21973766 #
3. danShumway ◴[] No.21973766[source]
Given the rulings around LinkedIn and public data, is it feasible for a competitor to scrape some of that information? Would they get sued to oblivion if they tried?

Not needing to start from scratch would make building a competing service a lot easier. Sites like Stackoverflow have reasonably open licenses on user data, so you could theoretically use that data and build an alternative if the site fell apart. I'm guessing that's not the case for Goodreads though, at least for things like reviews.

But even pulling in basic category information would be easier than starting from scratch.

replies(1): >>21985622 #
4. lkbm ◴[] No.21974724[source]
1. "Date Read" UX is terrible:

My family was recently shocked to discover that adding a book as read and rating it doesn't automatically set the read date. I don't think it needs to do so, but these people are computer-literate and have entered > 1,000 books a piece. And they didn't know how it works. That suggests poor UX.

As someone who DID know this and and always manually add Date Read...here's how to do it on desktop: hover over the shelf dropdown until a little popup appears above, move up to that popup without it vanishing, and click "Write a review", which secretly means "write a review or enter date read."

2. It's slow.

Loading just the html of the front page is > 3 seconds. Loading My Books is > 5 seconds. Again, this is JUST the html, excluding no js, css, and images.

3. Nav is bad

This combines poorly with the site speed. One thing I do most frequently is to look at my most recently read books:

- Go to goodreads.com (3+ seconds)

- Click "My Books" (5+ seconds)

- Click "Read" (It defaults to books read and on your to-read list all mixed together.)

- Sort the list by Date Finished (5+ seconds)

- Re-sort the list by Date Finished because it did ascending the first time. (5+ seconds)

(Obviously, I could just bookmark that page with the desired params, but if I'm bookmarking to avoid having to use your site navigation, that's a UX issue.)

4. The recommendation engine is bad.

Various people have mentioned this. I will grant that recommendations are hard. But basically, don't use the recommendations. Use the lists manually built by users. (But note that on most lists the top spots will be pointless recommendations that you read Harry Potter. Gee, never heard of THAT book before, thanks!)

5. Lists aren't super-accessible

As I mentioned, the lists are much more useful than the recommendation engine. They're under Browse->Lists.

There's a search at the top of every single page. It searches books and authors. Not lists.

If you're in the lists section, it...still won't search lists. There's a tiny search box on the lists page for this.

6. Search Breaks Middle-Click

When I search a book, it populates the results without me having to click through to the results page. If I want to open those results in a new tab, though...nope! It'll just re-open the current page in a new tab.

7. Their export tool doesn't work right.

This is a minor quibble--it's VERY nice that they let you export your data at all--but I recently discovered that a lot (most?) of rows in the export are missing the Date Read field even if you entered them. Not all though. I don't know what the pattern is, but it's annoying.

Basically, I think Goodreads has approximately one engineer, whose job is to do some tweaks for the marketing team as needed (they renamed Giveaways recently-ish). There's clearly no designer, as UX has been essentially touched in the 12 years since I joined.

It doesn't need a sweeping redesign, but there are obvious UX tweaks they could have made at any point in the past decade and instead didn't. And some performance work, please!

replies(1): >>21980465 #
5. yazan94 ◴[] No.21980465[source]
Thanks for the detailed reply! I admittedly am not a GoodReads user (I have an account but never use it) so I wasn't aware of all these pain-points
6. zem ◴[] No.21985622{3}[source]
i would love to see that happen! i have occasionally wondered what it would take to reboot a goodreads clone from scratch but importing the data is definitely a plan with a higher chance of actually replacing the site.