Heh, and now I still forgot to check that this is actually what resulted from my implementation - until just now, and nope! The 5 recent releases for which I haven't gathered a rating yet are at the far end. Luckily they've all been rated already, so that rating will be gathered in the next couple of hours. I'll fix the bug with the algorithm which places these there on Monday.

I see I managed to break the "jump to page" function somehow. I'll try to get that fixed first thing tomorrow.

Fixed. That was quite a brainfart which allowed that to happen, together with a near-complete lack of testing. >.<

This luckily wasn't a code bug, but just those 5 games having an uninitialized state for the rating.

So actually zero ratings on the store? Heh, shows the level of interest, not even knee-jerk low ratings...

No, these were games that were added in the time between when I finished my initial rating harvest, and when I pushed the code live. They all had ratings by that time, but I hadn't gathered them yet.

(I effectively update ratings twice a day for games added in the last month, and once a week for games beyond that, but that updating code doesn't run continuously on my local development environment.)

When I add a game, I initialize its rating (to "not yet rated", making it slot in after all the 4.0 ratings). For existing games, I provided an initial rating. But for the first few hours after I went live with this code change, these 5 games fell into the gap between those two states.

FWIW, there actually are 7 games like that (and hundreds of expansions and goodie packs).

I added a small (if not as small as I'd have liked) font file for the symbol glyphs which I'm using everywhere (all the little triangles and arrows and goodie icons), since apparently too many of them are missing in cases where I'd have expected people to have fonts installed which should've contained them. I'm curious if anyone here notices the difference?

I'm not sure, but I don't remember seeing arrows in the includes list, and now I do.

Hello there !

I've now used Gamesieve a bit, which is great at what it aims at, so here's my feedback, taking in account what could (in my opinion) earn it an extra star... and hopefully more recognition.

First, it's interface. It's fairly good, but sports a bit too much white and stark contrasts, without a darker mode, which make Gog's use of greys, black and white a tad easier for the eyes. At least at the age I'm nearing ^^'...
I didn't find the use of Germany's flag very intuitive to represent the Euro zone, but I've red the last five pages and understand your thoughts about it. I'd still recommend an Euro flag, but as long as one can pick the currency above, it only leaves minor differences of availability, I guess...
Then, there are other minor but noticeable graphical choices which my proofreading eye didn't find exactly excellent either, but it's obviously quite subjective. The filter category titles' positioning is a bit approximative above the filters, since the spacement has to deal with various size of arrows, (un)check marks, etc. Some of which being even a little misleading, like the merely graphical arrow in front of every filter... I'd ditch that arrow, move the one expanding the categories to the titles' other side, line up properly all symbols, and this way align all titles in a more precise manner. Bonus : it moves the actually interactive parts a tad closer to the middle.
This is only one example among others in your mockup : I'm not absolutely convinced by the positioning of your Search panel : it's neither exactly aligned with any information under it, nor as clean as it would be on the far left under Gamesieve ; I wouldn't center informations such as "Last year", "Previous year" (etc.) on top of one another and just align them (to their left) as you did everywhere else ; I'd suggest inputing the tags (and "last year", etc.) with their proper capital (first) letter ; I don't see the use for the arrow in front of game titles "included" in any pack, even though I guess you wanted an equivalent to the icons used before other informations (manual, etc.) and it's still a good way to mesh them all together ; the general organization of the central information feels a little cramped, but you managed to reach a solid compromise. I understand that you tried to regroup informations directly related to purchase on the left side, but I'd still, for one, favour the opposite solution : with tags on the left and the price on the right, leaving the historical elements in between, you'd probably reach a clearer state - on top of increasing available space for the (marketing) historical information. And, honestly, I'm not saying that because the result would be closer to Gog's catalog (thumbnails, etc.) ^^ ! It's just part of my experience at proofreading mockups...
There are other small details like those - I'd experiment with maintaining, in the right section (Available now, etc.) the same codes used in the middle (regarding development information above the game's title, etc.), but I don't want to spend too much time listing those.

I, for one, really appreciate the "Frequency of sales" filter... Very useful. That said, as others have mentioned before, placing all filter catogories related to sales (on the left) before any other practical information is completely debatable. If you intend to have a website actively used to be browsed, while discovering new games - rather than merely buying those one came for -, I'd rather have the opposite order... Of course, since one can always close the categories he doesn't need that much, it's not that important.

Regarding the search engine itself, as HunchBluntley said previously, I wouldn't underestimate the common user's need for simplicity, since most people don't use (even moderately) complex search commands when the most simple ones lead to good enough results. I'm surprised, for example, to get 26 results for "Quake" and don't quite understand, at first glance, how come games unrelated through any visible information can still be listed along.

In the end, in my opinion, the reason why this kind of alternative can take off or not doesn't depend that much on the said search engine - your lovely baby ^^ - and more to the amount of situations leading to its use. Everytime I, as a user, have still to manually open gog.com to browse for new release information, reaching Gog's blog, forums, wishlists (Dreamlist, etc.), even without any part related to my account, it's that much opportunities lost to consider your front-end. Any means to avoid that would be welcomed. Which is why the "Available now", plus the "Coming soon" sections (and any similar idea) hold a lot of value... and I'd bump the "Privacy" disclaimer under them. You may soon have to add additional details of this kind anyway, at the bottom on your page. Long story short, it's still a great idea backed up by impressive work, so I've advised people around me to give it a try... And yet, at least for now, I've only been thinking about it in a few niche situations (precise statistics, semi-unknown titles hard to find through the common search engine, etc.). More associations would be required to make it a permanent daily stop...

Hi! Thank you for the extensive feedback - I really appreciate it, particularly all the detailed comments about graphical layout. I have soaked up some knowledge in that field which would make me from a decade ago marvel at what all I pay attention to now, but am far from a being a visual designer myself, so there are definite limits to how good I can make things myself.

I'll respond to a few specific things below, but I'm taking everything to heart. A lot of the things you surmise are correct, and a lot of the things you suggest I've tried and/or don't do for deliberate reasons - but I'm going to have a second look at the validity of those reasons. I'm probably not going to change anything about the layout until I'm done with (or temporarily postponing, or giving up on) the next main feature I'm working on, but I already realize that once that's functional, I'm going to have to rethink some aspects of the layout anyway, so will definitely revisit this at that point.

GameSieve does have a dark mode, which is automatically applied when you tell your OS that you prefer dark mode. (In Firefox you can also do this in the browser: preferences, general, language and appearance, website appearance, dark). All my attempts at going for a grey background ended up looking like a website from the 90s, and I'm not willing to sacrifice the information density to add lighter colored boxes on top of that the way GOG does - but yes, I do know that many people are using monitors which are far too bright, and that the light version of the site is not ideal for that. I'd be interested to know if dark mode is the solution for you, and you simply weren't aware of it, as I've been pondering if it's worth the complexity to add a manual toggle for it.

To be fair, it really only represents Germany. Most of the games for most of the eurozone have identical prices, but GOG really does distinguish on a per-country basis, so I need to do that, too.

The arrow does serve a purpose, in that on smaller (but not mobile) screens, and/or when people have increased the font size, quite a few filters wrap to two lines of text, and some "list item" indicator then becomes really useful for differentiating between distinct filters. But yes, since I added the triangles to the filter titles to indicate that the filter blocks are collapsible, I should at least try to line out all the text and icons.

The left edge of the input field should (and does for me) line out exactly with the left side of the "box covers". However, I see on a screenshot I have from a Mac user that indeed they don't do that there, either. Thanks for calling this out! I'll need to look into what's causing that mismatch. (edit: Urgh, font metrics again... Luckily I see a clean workaround, so that at least I'll apply tomorrow.)

As you note, you're not the first to mention this - but the general way you phrased this was a really big drop in the bucket that's made me re-examine what I can do here. Technically, I could split up my search into a title search, and only do an "everything else" if there are no hits on title (and/or show those hits collapsed-by-default, or put a line in between (but that doesn't work for different sort orders), or ...). However, the usecase which I keep coming back to as justification for what I'm doing is a search for "RPG". There are 7 games with "RPG" in the title, and another two with "role playing game". Yet only showing those 7 or 9 games on a search for "RPG" is clearly not the correct answer. Now RPG happens to be a tag/genre as well, so for that particular usecase I could do as GOG does and only search through title + tag/genre, but there are many similar cases where those wouldn't contain the words people would use.

Anyway, I'm hearing you, and although I'm not actively working on changing this just yet, I'm very much trying out different solutions in the back of my mind to see what would be necessary to make this better.

Fair point! I don't want to hide the privacy section away, particularly now that I've added affiliate links, but yeah, once I add user-specific information, I'm going to have to write a lot more. I'll probably start by making that section collapsible as a first step, and/or see about splitting it into a shorter actionable section at the top, and a separate page (?) with more details.

These two things should now line out better and as intended, respectively. (I'm not entirely happy with the white-space around the icons for the filters now, but still think it's an improvement.) If you have other cases where you see things not lining out, but where they could without re-ordering the elements on screen, I'd be happy to hear about them, since it's quite possible that I made more bad assumptions which are invalidated by font metrics.

Suggested feature:
When listing historical prices, please consider also listing the discount.

I consciously don't, to not make the information overload even worse than it already is.

What is your usecase for wanting this? Insight into base price changes? (I'll probably eventually do a foldout with full sales history listing.)

For those of you who actively used GOG mixes back when they existed: How important were the comments on each game, versus just the fact that a game was present on a certain list? If I reimplement the feature, but leave off such comments, what percentage of the value of gog mixes would still be present?

For some GOGmixes, the comments attached to each individual game had provided not only justification for their inclusion on said list, but, often, had contained useful information, as with the following examples:
GOGmix: GOG Games with Serial Numbers
GOGmix: Games You Probably Should Not Support

GOGmixes without comments are far less intriguing, as, apart from the title of the list itself (and possibly star ratings), no further context is offered. Example (not intended to criticize the author):
GOGmix: Top 10 Puzzles

Generally, for the GOGmixes that I had followed, approximately 25-30% of original perceived value would remain.

Edit: By replacing : with %3A, circumvented automatic GOG forum omission of Wayback Machine portion of each URL.