Unfiltered observations and species search results slowing iNat speed (Explore, Identify, Suggestions)

If I search a taxon and location in Identify it returns numerous “pages” of observations (the numbered pagination buttons at bottom). I sometimes get an error message if trying to click on the very last numbered page from the first page. Also, when I login to iNat it initially opens to Explore showing all global observations of all species. Similarly, when I’m in Identify and select a location alone first (without taxon) it initially shows all taxa for it, or if I chose the taxon first it initially shows observations of it globally, both of which are often vast numbers. Also similar, when using Suggestions after a taxon is selected it shows numerous global species if I clear the location filter. Could the way the website often shows all observations (in Explore and Identify) or species (in Suggestions) “by default” be slowing site speed?

Update: it was since suggested that they do. One potential solution to consider would be to not make Explore, Identify, or Identify Suggestions not show their respective results until both taxon and location are specified, as otherwise we’re loading all global observations (e.g. Explore). Would this work, or any other ideas? I also was just wondering if these causes of slowness were widely known in general. This topic was originally a comment split from another topic.

Now you can’t click from the first page to 100k+ observations page as I remember, so error message pops up when you click too far (but you can use filters to show in reverse order and it will show you old observations).

2 Likes

That’s a good tip for that part of my question. Do you know if the rest of what I asked is likely or has been discussed in the forum before?

I think it could be that, though the biggest problem is in iNat structure itself, here I listen to letures and GBIF is often mentioned on how fast it shows you much more results than iNat has with no problems, it’s something about tables, etc., I think that’s what staff is working on recently, with the post looking for a new position hinting on that.

2 Likes

yes, but this is generally unrelated to inability to get high-numbered pages. for more information about pagination, see https://forum.inaturalist.org/t/unable-to-load-older-pages-beyond-page-100/13139.

1 Like

Okay, the getting to the last page aspect itself doesn’t really bother me much anyway. What about how I also wondered about Explore opening to show all global IDs, and that Identify and Identify Suggestions sometimes seem to “by default” open up to show many/all obs.? I don’t mean it needs any immediate solution/change, just to note it to keep in mind if potentially affecting speed. For example, I’d wonder if it would be better to require users to enter taxon and location (maybe to select both) before the system would show any observations, whether on Explore, Identify, or Suggestions.

we’ve already answered your question:

how to go about making the system more efficient is a complicated question that is probably impossible to discuss properly in the forum. if you want to try, you should probably start a new topic.

Fair enough, although there seemed some uncertainty before your last comment. Also, when I once raised that topic “Improving site speed” and asked about various aspects of the site which could be impairing speed, some members (I forget who) seemed to cast doubt that those kinds of things would be. But if they are and it’s complicated, that’s fair enough and basically what I expected. Lastly on this topic just to know, I notice you often reply about website issues, so am I correct to think you’ve worked on the website programming, or you just know about those things?

Default is definetely set to the “world”, so I make sure to fill in info as fast as possible so all of observations don’t load, it’s especially valuable for explore where it won’t even show you the correct info about the latest uploads. So if you’re worried about it causing something that’s the way to shorten the big load. (I also often do search for my own obs first, then unmarking it)

1 Like

I agree. I was also more referring to how since it’s doing that for every user by default (at least before they filter the search) it may be contributing to website slowness. So I’m suggesting if that were changed more “upstream” it would improve website speed for everyone. As for my own usage it’s not a big or new problem, although the site speed is a challenge many recognize and which hopefully will improve in time.

1 Like

i don’t see this, but it’s worth noting this post from that thread:

(to me, this sounds like a locking / blocking problem.)

also, marina_gorbunova already mentioned it, but for anyone who thinks they’re qualified and really wants to poke around the system to improve performance, you should look at this iNat blog post: We’re hiring a Systems Architect to help iNaturalist scale! · iNaturalist.

1 Like

I agree with the comment you quoted. There were a few other comments there that somewhat questioned whether there were slowness issues at all or whether certain proposed causes were actual causes, and one of your first replies mostly only stated that causes needed to be specified more clearly without confirming any in particular that I mentioned. I also mentioned the specific causes I listed in the current topic vin comments on other topics long ago (forget where) but they were never confirmed, which is why I asked again here. Note that someone not engaged in site programming can only ask if each cause is contributing without initially knowing. I was also asking here not only as a question, but given that the causes do contribute to slowness to simply raise the concept to anyone working on the site in case it wasn’t already known. It may well already have been known or anticipated, but ensuring that was why I wrote this. I haven’t yet set out to learn programming beyond basics. Your reply above “You question has been answered” seems blunter that you respond to at least most other members, but more importantly seems to intend to stifle or end discussion. You also made similar comments when I noted map location search problems in another of my past topics, except that was an actual problem which you contested initially. Given that the location issues I raised were actually real, it also explains why members seem to raise location issues regularly now on forum independent of me.

I above asked if you work on iNat programming (which you didn’t answer) not to criticize iNat or whoever’s working on it including you, but because you seem to reply to topics/comments about speed or location issues as if under personal critique. It would be better to just let people discuss these things. We technically don’t know how much of what I’ve mentioned is widely acknowledged or if alternatives have been considered (like my suggestion to not let the pages show taxa until location and taxon has first been selected). I’d be more interested in hearing perspectives from more people instead of having to assume “question answered.”

1 Like

when i look out my window down the block, i see a big oak tree with mostly bare branches swaying in the wind. it looks like there are buds, but i can’t tell if they’ve broken yet because it’s kind of cloudy. so it’s just dark brown against a smoky white. earlier, there were woodpeckers pecking at the tree. one of them had a bright red patch on its head, which i could see even at this distance against the brown and white. but even if hadn’t seen the birds with my eyes, i heard them making their distinctive pips, and i knew exactly what they looked like. some people don’t like woodpeckers pecking at their trees because they think they damage their trees, and surely it would be tragedy if a big oak branch fell onto someone’s head. when a branch falls, will it be because of a pair of woodpeckers in love? will it be because of the tasty bugs who love the tasty bark? will it be because of a hurricane that blows through? or will it be because of a human who loves the feeling of power while wielding a chainsaw?

6 Likes

What’s this about, or was this intended for another forum page?

2 Likes

@benarmstrong @naturalist_aditya @ivanmunks @zdanko Care to explain the meaning of pisum’s post you liked? Looks like an insult.

@brian_d – no need to assume sinister motives. there are no insults in my last post. my last post was just me taking a moment to connect with nature and sharing my stream of consciousness. i assume other folks liked it because they appreciated the shared moment.

5 Likes

Why would your reply to my comment be to share your stream of consciousness?

1 Like

sometimes it helps to take a break

2 Likes

It seems implausible that you would reply to me with an unexplained stream of consciousness ending with “or will it be because of a human who loves the feeling of power while wielding a chainsaw?” It also seems implausible that others would assume you’d reply to people with unrelated content and that everyone will somehow think that makes sense in the way you claim.

1 Like

ok. i’m going to excuse myself from this thread altogether. if you think my post was insulting, feel free to flag it. see you elsewhere in the forum.