IT Development Plan 2019-2021

nadia · January 17, 2019, 12:26pm

This is cool and could be very helpful for the different topics @noemi see https://docs.google.com/spreadsheets/d/10bdjvFshsJSkn9dsxPSe5xEl8eg760YqzMLPSJd0Y7Q/edit#gid=0

matthias · January 17, 2019, 3:58pm

Let’s collect the requirements for the façade websites for POPREBEL and NGI Forward

In the plan above, I propose this item to create the façade / funnel websites for both POPREBEL and NGI Forward:

@owen has some nice way of creating such things, so let’s collect the detailed requirements for it so that we can contract him for that and he can get to work.

Outline of things I already know or decided

The façade websites will have their own domains and serve as the first point of contact with the POPREBEL / NGI Forward project.
We want the façade websites to be single-page websites. If they have a menu, the menu items should scroll the page down to the corresponding section.
We’ll collect requirements for the content of these sites, but leave the visuals / design to Owen. There may be a few exceptions to that, where needed.
We can (and will) pull live information from Discourse and embed it on the façade website. This can be recent posts, statistics like numbers of likes and comments, a list of contributors etc… For inspiration / examples of such content, see Owen’s previous work at participio.netlify.com.
Self-hosted. Unlike Owen’s site for Participio which is hosted on netlify.com, we want to host the POPREBEL / NGI Forward façade sites on our own server. Should not present an issue, as netlify.com is only a deployment platform, not locking in development to that platform.
No server-side scripting for page rendering. One options is that the façade website would query the Discourse API using client-side JavaScript code to obtain the featured content etc. – but that puts unnecessary load on Discourse. The other option is to use a static page generator and let that run regularly (once per hour? once per day?) on our server to re-generate the static façade websites. (Does any of these options work for you @owen, and which one?)

Open questions

Which main sections do we want on the POPREBEL façade website, and which on the NGI Forward façade website? How to call the menu item for each section?
What live content elements and statistics should we pull from Discourse for the POPREBEL façade website, and which for the NGI Forward façade website?
What introductory content do we want on the POPREBEL façade website, and what on the NGI Forward façade website? Note that this should be very short content pieces (a paragraph max, plus image / video / animation).
Who should Owen contact to write that introductory content in each case?
What are the exact requirements by the NGI Forward consortium about the integration with the project’s current website (which is a point in the application)? Does that affect the requirements for the façade website collected here? Or, if any of these is not known so far, who do I have to contact about it. Probably that’s a question for @anique.yael.
And for @owen: Do you need any additional information to what I asked for above in order to be able to give us a price quote for these two façade websites? Assume complexity to be very similar to what you created for the Participipo project.

Can people please give me their input on the above questions? Just pick what you can answer or contribute to, please – @nadia @johncoate @noemi (@alberto). Thank you.

alberto · January 18, 2019, 9:50am

From the phone and in a hurry, but: project websites are an EU obligation, so subject to certain standards (logos etc.). They fall within the remit of the coordinator. We cannot just make a move and build websites that make sense.

Two possibilities:

Negotiate with the coordinators. This should start very soon, I believe in POPREBEL the website is due at month 3.
Leave the institutional websites alone, and build onboarding funnels instead, as we did in the past.

anique.yael · January 18, 2019, 11:09am

Thanks @alberto and I agree although not completely across all the technicalities. As far as I see it, project coordinators are responsible for the project’s websites - and yes @alberto POPREBEL’s is due for Month 3, the specs for which are already under way I believe. The idea is that we can pull from there and aggregate to there. But perhaps I’m not clear on what you mean by a facade website - do you mean more a facade for all our fora and discussions as per [quote=“matthias, post:22, topic:9202”]
This can be recent posts, statistics like numbers of likes and comments, a list of contributors etc…
[/quote] ?

Something to keep in mind is that we are still in discussion internally and with some partners around how to go about the multilingual component and accessibility - for example: do the Czech audiences go to a Czech only landing page or do they go to one with all fora in all languages? This came up in POPREBEL and will quite possibly be different in NGI Forward considering the different nature of the project.

NGI Forward is a sector-building and stakeholder engagement project for a high profile initiative, with a whole other project taking care of communications and media. We will have to work very closely - particularly @nadia - with the coordinator Nesta and the communications project (name not yet known) on all our comms. I’ll bring up your question in the kick off meeting Monday-Tuesday and get back to you.

matthias · January 18, 2019, 11:35am

That’s what I mean by the “façade website”: we’re not dealing with the institutional website at all but simply need a pretty-looking overview site where people arrive and are then guided to our actual, Discourse based discussion platform. Since we decided above to host the POPREBEL and NGI Forward discussions right on edgeryders.eu and not as a federated Discourse website on their own domains, a façade / onboarding funnel website is the only way to add more customization, own domain, some branding and a proper sense of what’s going on to the first page people see when visiting an invitation link they got.

That should also answer your question @anique.yael what I mean by “façade website”. As an example, Owen’s participio.netlify.com is a façade website for the Participio category here on edgeryders.eu.

To and from where, the institutional project websites for POPREBEL / NGI Forward? If that implies automated data exchange between our Discourse based discussion platform and that website, please provide the tech specs for that. Didn’t hear about that requirement so far.

As far as our Discourse based platform is concerned, that’s just about sharing a link to a language-specific sub-category vs. the link to the top-level category. But Discourse is not about landing pages, so that’s a decision that will be needed for implementing the façade website(s), which is meant to provide the landing page(s).

Means, we’d need a rough / basic decision about this quite fast (a week?) so that Owen can get to work.

Thank you. My question is especially about how much the look and feel has to be the same as that of the existing NGI Forward institutional project website and what else is meant by “integration”. If a link that goes directly from their website menu to edgeryders.eu/c/categoryname is not acceptable (because the look&feel and domain is different) then I would propose to link to the to-be-developed single-page façade website instead. It would still not be on the same domain as the institutional project’s website. If they want that, it’s better that we develop the Discourse façade website within the same framework and code base they use for the institutional project website for NGI Forward.

anique.yael · January 18, 2019, 12:01pm

Got it thanks @matthias! RE: multilingual architecture / landing and [quote=“matthias, post:25, topic:9202”]
Means, we’d need a rough / basic decision about this quite fast (a week?) so that Owen can get to work.
[/quote]

I defer to @noemi and @johncoate here. No doubt they’ll be in touch to pull you into the discussions. NGI Forward may take a little longer but I’ll prioritise feedback from the consortium at the kick offs.

noemi · January 18, 2019, 3:09pm

Exactly!
What I can report back from these training days is that most partners and our comm mgmt team is ready to work with a big POPREBEL category + subcategories per countries, as well as anything that we propose i.e. different POPREBEL [country] category + subcategories.
No strong questions or opinions were voiced about language and web design.
Happy team over here.

matthias · January 18, 2019, 3:17pm

Good. In that case, I’ll simply decide for POPREBEL as follows:

There will be language-specific versions of the façade website, but these will simply be translations of the English version, with links adapted to point to the respective language-specific spaces of the POPREBEL discusson on edgeryders.eu (however our community builders decide to structure the space on edgeryders.eu).

Links to share to the façade website would look like this, using an example domain name:

http://poprebel.eu/ - English version, the default
http://poprebel.eu/cz - Czech version
http://poprebel.eu/hu - Hungarian version
… and so on

Not all of the translated version might be there on launch day.

noemi · January 18, 2019, 3:46pm

Give me until Mon/Tue to process all this and gather thoughts post-training?
But it seems reasonable… it’s just that I’ve seen already many threads - riot included, about the languages, new features for preventing trolls etc. Need to process a lot.

If urgent, dont wait though!

matthias · January 18, 2019, 3:50pm

No problem, it has around a week from now. Just saying that, if I don’t hear anything to the contrary, the above is what we’ll use when Owen starts to implement the site.

Where I’ll need your input is about the required content elements for the facade website (see the detailed questions above). But again, take your time to think about it.

owen · January 21, 2019, 8:43pm

I see three options:

Client side call to the Discourse API - this is probably the easiest and guarantees that all content is up to date on the site, I don’t know how resource intensive it is on the server side but loading the data asynchronously happens fairly fast, using the Vue architecture I used for Participio.
Alternatively we can use the great script @Hugi developed that calls the data at fixed intervals and perhaps caches it for a certain amount of time so it reduces the number of calls on the Discourse platform. This would be essentially the same as the first option while moving the request to a different server.
The third option would be as you suggest to import the data at set intervals and generate the pages using a static site generator. Nuxt, the framework I’ve been using has the ability to do this very well: it will store the response of the REST request, pre-render the site during build and serve everything including content as static files - which gives us the benefit of fast load times and is great for security (no Wordpress type vulnerabilities, etc).

I would lean towards the third option, generating the site at a fixed interval that makes sense for the content being published.

Do you need any additional information to what I asked for above in order to be able to give us a price quote for these two façade websites? Assume complexity to be very similar to what you created for the Participipo project.

I would need to look at the content to get an idea of this, but I imagine it would be similar in scope to the Participio work, although the script for storing the content and generating the site will be different from the code used on that site if we go for option 3.

I would add that enabling Single Sign On (SSO) on these sites, so that users can stay logged in and reply to threads would be a very good idea and encourage contributions to the platform. Allowing new visitors to sign up to the platform from these sites would also be great and tie the platform a bit closer to the content.

I saw there is now a dedicated sign in page at http://communities.edgeryders.eu… does this use SSO to keep users logged in across different directories? If this is the case, it would be fairly straight forward to implement. Users will be aware of their logged in state, and it will make it easy and transparent for new visitors to sign up.

Means, we’d need a rough / basic decision about this quite fast (a week?) so that Owen can get to work.

I’m available to begin when there is a content brief and the requirements of the organisations are confirmed, which sounds like something still being discussed? As soon as I have some basic idea of the content (and scope such as different languages) I can provide a quote and start. The next month will be ideal on my end as I have a bit more room to breathe to start something new, while continuing work on the Participio page.

noemi · January 24, 2019, 12:41pm

Thanks @owen. This for me and @nadia to discuss, start building the source content and get in touch with you soon. The work to happen in Feb - sounds good.

owen · January 26, 2019, 5:09pm

Great

matthias · January 30, 2019, 10:21am

From chat, putting it here to not be forgotten.

Noemi mentioned that “[the façade website] will only come later, as the launch is to be done much low key as far as I understood preliminary from Nadia”.

To which I said:

The idea is that the façade website is a more pretty intro for users before they proceed to the Discourse platform. You don’t have to mention “populism” there at all, or the EU project. It’s not the institutional website of the project, but our own usual “engagement funnel” website.

If you and Nadia still not want it at launch, that’s fine with me, but note that the ways to customize the look and feel and layout in Discourse are limited (and there’s no way around that given the IT budget as it is).

And adding to this here:

You (@noemi and @nadia) can please directly coordinate with @owen about when you want what kind of façade website ready for POPREBEL. If you want to have it ready for the platform launch in early March, please make sure to give him the requirements in early February. If you want it ready later, not a problem with me. Just keep an eye on this please

owen · February 6, 2019, 3:55pm

@matthias @noemi @nadia - any updates on this?

matthias · February 6, 2019, 4:00pm

Only that I have heard from @noemi that they decided to launch the façade website for POPREBEL later, not together with the main platform in March. Noemi can elaborate about their time plan and when you can expect to get the requirements for your work.

noemi · February 7, 2019, 8:05am

Hei! indeed, we will postpone this to a date probably closer to the European Elections, when the engagement story will be clearer. For now the consensus is that we wish to keep things simple and offer (only) a dedicated safe space on the platform for the project - safe for participants, but also safe for the team.
Makes sense?

thanks for reaching out, and of course we’ll be in touch at least a month before any release date to see what your schedule is. how do march/ april look like for you @owen ?

owen · February 7, 2019, 10:04am

Hey Noemi, thanks for the update - no problem for March/April, as long as we can touch base over content & marketing requirements before then.

matthias · April 16, 2019, 7:39pm

For our ethnographer @amelia and for @alberto: I have made a proposal for implementing the Open Ethnographer multimedia coding feature in the first post above (under bullet point “Multimedia coding”).

Please have a look and either ok the proposal or propose changes. Specifically note that I propose to code video and audio content only by time, not by time and video frame section.

alberto · April 16, 2019, 8:05pm

The reason for this is self-evident. However, it poses a problem in data archival post-project. How would that work? Could we archive video raw data + annotation?