Open j08lue opened 11 months ago
Lighthouse parses out the rendered page to see if DOM is structured well enough for SEO engine, and I am glad we are following a good practice on it. However, this is when the search engine already reached the page. Dashboard is currently set up as a single-page application, meaning there is no page for search engine to crawl without extra setup. I wrote about this issue a while ago : https://github.com/NASA-IMPACT/veda-ui/issues/11
👉 In short, we don't have a lot of options to optimize SEO without shifting the foundational framework. But this might be a good time to think about how we can gradually change the framework, combining it with other issues (ex. submodule structure not being able to solve the problem of 'forking' the repo.) that we have observed so far.
AAt least I am seeing a better score from the cache policy that Abdelhak put on the staging. But there are some items that we can do more, such as: (I will add more to the list if I find more.)
Did we ever get more information on why performance was so low here? or has that score changed?
Also checking in on this ticket... the report can be found here. Looks like the width and height of images could be made explicit to help improve the score.
Thanks for sharing the report @jerikac it's really helpful. Looks like the overall performance score has vastly improved from the original report. Here is an updated image of the scores and the top opportunities
Yep - I think the goal for GHGC is to improve the report to =<85
@jerikac @Jeanne-le-Roux @siddharth0248 looking at the lighthouse report again, two low hanging fruit I can see are:
I think we could take a quick look at #2, could you all take a look at the first one?
A check of the Lighthouse score showed low performance and high SEO scores. We would probably expect the opposite to be true...
We should look into this and make a good analysis of the status and optimization potential
Acceptance criteria