-
Notifications
You must be signed in to change notification settings - Fork 189
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
LearnOSM need to be tidy #306
Comments
You could have a look to the result : http://pyrog.github.io/learnosm-sandbox/en/ Note: the german page about is missing to check what happened when a page is untranslated (see #252) |
Hi @pyrog if I understand you are saying "tidy" to mean two different things?
I don't see there being a lot of different simple moving the files out of the "_posts" they are already in there in individual language folders. There could be better organization within there. As for the resulting site you created I'm a big confused, it looks like a default Jekyll site. The LearnOSM site as it was done now was created with a design process and I think is a modern site and a good look. |
Of course, I use the default template 😉
|
Hi @wonderchook It is also an available sandbox where we can test anything around structure.
Generally (in issues+sandbox+discussion+lists...) we would like to assess the pros and cons of blog structure and regular structure. As it stands with the _posts folder, category organisation, dated and ordered filenames... It looks slightly odd and we are trying to understand the current implementation as blog. My feeling (for the time being!) is that it is not completely suited and requires a bit of shoehorn. I would like to know more about the history and previous choices. I would love to get critical feedback on any structure. |
With the suggested design, it will simpler and obvious 😄 :
So currently we have for the file His jekyll source filename:
His frontmatter: ---
layout: doc
title: Pour aller plus loin
permalink: /fr/beginner/moving-forward/
lang: fr
category: beginner
otherguides: "Les autres niveaux"
--- After the tidying of Mr Clean 😉:
Is front matter will be: ---
layout: doc
title: Pour aller plus loin
category: beginner
otherguides: "Les autres niveaux"
--- |
Tidying images: In the images folder, many images are redundant:
![paper scanned][]
…
[paper scanned]: /images/en/beginner/06_field-papers/en_beg_06_field-papers_image20_watchsnapshot.png ![image](/images/fr/0300-12-23-private-data-store/image19.png) With the regular expression below, I found only 1 796 images used by .md files in !\[\w*\] The regexp found images like: ![image20][]
![]({{site.baseurl}}/images/en/intermediate/en_int_ch6_image14.png) If I search only in the Finally, there is 3 036 images in the folder |
@pyrog thanks for your work on this. I'm no expert, but reading through what you are proposing, it looks as if we would have a far easier to understand 'filing system' for the files needed to display the site if we adopted it. I'd be interested to see what the proposal would be for a transition - How easy would it be to change from one filing system to another, and would the site be available throughout? The site is constantly being updated, so a transfer would get a little complicated. I'm guessing we could branch, change the structure in the branch, alter the file which shows the web address, and if there is a problem quickly change back until it is sorted - the proposed site can be viewed on jekyll when being 'built' - need the advice of someone with more knowledge though. My own level of knowledge is such that I would not attempt to start something like this proposal unless there were assurances from users such as yourself who understand how this should work, that they would stay with the project & help with ironing out any glitches. Whatever system we end up with needs better documentation of 'howto' for those who are willing to write articles about mapping, HOT, etc., than we currently have. For the future, and those who follow on from us, it would be good to have a more logical system so mappers who wish to help others can easily write articles & submit them for inclusion. Could the change over be managed incrementally? I am concerned that the amount of effort involved could outweigh the benefits - I need someone with more knowledge on the subject to offer advice so we could then discuss & decide what to do. |
Where you want inside the project tree 😉
I just made a test and it works 😄 In ![A village in Indonesia](images/village-in-indonesia.png) But if you want to use the same image from the french translation, it could be tricky:
The simplest solution for editors is to choose the first one, but dev need to:
Example for case 2. In ![A village in Indonesia](images/village-in-indonesia.png)
Currently, I use the KIS method : If an image should be localized, I suggest to add the language code in the prefix: |
Yes, it's easy with git and github |
Tidiying images see discussion in #94 |
I updated http://pyrog.github.io/learnosm-sandbox/ PS: type "iD" : it find 2 pages with the same title. They are in 2 different guides 😉 |
@pyrog - the proposed simplifications sound great at first read. Would be awesome to see a pull request against this repo. |
@lxbarth If we agree to make this switch, what about these steps:
|
Work in a fork right away, pull request as soon as you want a specific On Tuesday, February 10, 2015, althio notifications@github.com wrote:
|
@lxbarth That is why @pyrog worked in a brand new repo, not a regular fork. Based on that, could you have a look at, comment and review: |
Hey @pyrog! I am excited about the prospects of changing the site structure to better fit the learning materials. I am interested in better understanding how removing the _posts folder affects the site. I tried to test the existing sandbox site with the "next & previous" buttons from the code I used in issue #270 but it doesn't seem to be working at all and I am not sure why. Any ideas? Also, above you mention taking out the permalinks and leaving categories but that's not reflected in the sandboxrepo, has this just not been updated yet or was there another reason for this? Here's the code I used for the next/previous buttons (same as in #270): |
Copy & paste of a summary comment from @pyrog concerning a filing system for images on potential new site discussed within comments on pull request #327 In "my" sandbox, I use only one image folder but there will be to many images. But I prefer to avoid language subfolder and use a language extension (See Content negotiation concept).
If an image is "neutral language" (like a picture of a mapper using a gps receiver) all "translations" use the "english" image ( But if the image contain important text or have a cultural meaning, for example french translation of a chapter use |
@jmarlena {% if page.previous.url %}
<a rel="prev" href="{{page.previous.url}}">← {{page.previous.title}}</a>
{% endif %}
{% if page.next.url %}
<a rel="next" href="{{page.next.url}}">{{page.next.title}} →</a>
{% endif %} It's possible to use a plugin like Jekyll-pagination but I worked on a TOC plugin. See pyrog/learnosm-mm But the current issue is how to use Jekyll plugins (or other static web pages generator) with github-pages. Do you have some tricks from Travis.CI team ? |
@pyrog I will ask the Travis.CI team if they have any different ideas/input on how to use custom Jekyll plugins with github-pages. I will also try experimenting with this. I am not sure maintaining multiple branches for this site is feasible with making learnosm.org more user friendly. The prospect of switching over to another static site generator seems more and more feasible to me. Also, I presume that TOC = Table of Contents plugin? |
@jmarlena & @pyrog for your info - we have a new intern starting with us shortly - not sure if they already have a username on github. @althio & I have put together https://hackpad.com/Outreachy-LearnOSM-content-ecmiGI21ZlM in an attempt to get Anwar up to speed as quickly as possible. I've started to update the files & images in line with the agreements so far, but there is a lot to do - any help appreciated. We may / will need to start allocating which files will be updated by which person to avoid us wasting time by working on the same file at the same time. As we have someone translating the site from English to Swahili, I've concentrated on making sure the English section, starting with Beginner, is up to date before it is translated, so that the new files in the Swahili section, and the images associated with it, are already in the correct format. Any help / suggestions appreciated. |
I added a comment on the hackpad under the comment section on UX resources. My hope is that the UX terms and resources I shared will help the new intern find relevant help and a framework for understanding the site's problems and possible solutions. For example, here's a learosm_site_audit google docs spreadsheet I started from a Content Audit template, which I originally found from this helpful UX Recipe Checklist. I think using google docs is a good way to manage content inventory/audit. I also added some comments on the hackpad summarizing the technical challenges I encountered with the site and some possible solutions. |
@jmarlena |
updating file in line with #306 and amending file so images will display
… removing redundant images
…on & updating image location as hotosm#306
amending image location in Osmand2 section in line with #306
@jmarlena @althio @lxbarth @pyrog I've added a second sheet to the spreadsheet at learosm_site_audit which shows the progress of the 'tidying', as well as some anomalies on files present in different languages. As I've been working through I've often renamed all the files in a folder so that it is easier to make additions later (only while we are still working with the '_posts' blog system) Basically I've spread the 'dates' so we can more easily see what is needed, and what we have. I haven't updated the wiki, as althio & I have set that as a target for our intern. I'm also hoping for help with the tidying. Although the 'tidying' is time consuming, I've made considerable inroads on the number of deleted redundant images (I'm checking that none are 'localized' before doing so), and making other slight improvements as I go - correcting broken links, adding a comment on the top of 'old' files that a newer version exists in another language, etc.. My intention at present is:
|
@pyrog I'm closing this now but I agree and it's a long term aim. Using our current system we' re gradually improving the situation. I'll keep working on this even though the issue is closed - trying to tidy the issues as well! |
Hi,
One of the big problem is that this site is a… mess 😉
It use essentially _posts but this is not a blog, it's a book. We don't need to put chapters in _posts.
So the actually structure of learnOSM is too complicated, difficult to understand, translate and maintain.
We should tidy it and separate better data and code:
Examples:
(so we should have the source image and separately the anotations written on them)
The text was updated successfully, but these errors were encountered: