linkUpdate #2 - July 2020

About two months ago, we announced coding.blog. Following a positive response by the community, we started the alpha phase of the platform. We released the publishing tool-chain and started a slow roll-out of blogs on coding.blog. We published the first update and outlined a plan for the platform.

Here's a summary of what has happened since then and what our future plans are:

• About a 100 blogs were activated for creators who signed up on our Prospective Creator's List. As mentioned in the last update, we opted for a slow roll out purposefully to be able to tend to the feedback of each new creator. As the product further stabilizes, we will increase the rate at which new blogs are activated.

• Progressing towards the planned beta stage, we have released the first public issue of coding.blog. We have a steady flow for publishing such issues on a recurrent basis, and as the flow increases, we will progress towards paid personalized issues (same format, personalized content).

• Based on creators' feedback, we've had lots of updates to coding.blog, CODEDOC, and the associated plugin. These improvements include:

  • Realtime feedback on publish process on coding.blog UI
  • Improved local preview, with fast iterative rebuilds and docker support
  • RSS support
  • New features in code snippets: linking to lines, highliting segments of a snippet with URL hashes, marking warning/error segments in the code, etc.
  • SEO and general performance
  • Lots of bug fixes

• We expect to progress according to our outlined plans, except perhaps we might roll out the curation system earlier than the tipping system.

linkPublic Issue #1

Our main plan for monetisation coding.blog is through paid curated lists of content, as mentioned before. Progressing towards that goal, we have published the first public issue. This issue which is a free and public collection of interesting recent articles about programming that we have hand-picked ourselves along with some community recommendations.

We envision our paid issues to have a similar format, but instead with personalized content. At the moment, we have a content flow that can sustain the public issues, and as the flow increases, we will move towards the personalized issues as well. We will keep publishing the free public issues even after personalized issues are released.

The first issue is comprised of articles mostly published on coding.blog, though it is not limited by that criteria. The quality of the delivered content is priority number one, so if there is any interesting article that we think our readers would benefit from knowing about, we want to sread the word to them, regardless of the tool-chain used to publish the content.

Since we can guarantee a higher baseline quality (in terms of readability, performance, etc.) on coding.blog, alongside provide more value to authors of articles published on the platform, we do prioritize articles published on coding.blog to some degree.

linkProduct Updates

Based on feedback and in collaboration with initial batch of creators and testers, we have fixed lots of issues and inconviences on coding.blog, CODEDOC and coding-blog-plugin. These updates include, but are not limited to the following:

linkPublishing Process

The blog management page now provides realtime feedback on the status of a publish job in addition to notifying you via emails. The more verbose error reporting of that, alongside the newly added knowledge base, should allow much more streamlined troubleshooting publishes.

linkDev Environment

Initial testers on the alpha stage were quick to point out that CODEDOC takes a long time to rebuild for huge blog repositories on a local preview. In light of that, we shifted to a tiered build process that builds component codes separately than markdown files, drastically speeding up the rebuilds (from reported sometimes ~17 seconds to less than one).

Additionally, we realized the setting up CODEDOC on docker on windows (for local previews) is not as streamlined as it could be. To fix the issue, we included a docker file alongside our boilerplate projects to help windows users avoid such inconveniences in the future.

linkCode Snippet Features

Functionality of CODEDOC's code snippets have been expanded. You can now link to any line of any code-snippet (it has a unique URL), or even to arbitrary selections of lines. When navigated to a document via such a URL, the selected line(s) of the code-snippet will automatically be focused and highlighted. This means authors can also use this feature to change highlighting of a particular code snippet interactively on the page.

Besides links, we added four new possible annotations for codes: Added and removed lines, which highlight the marked lines but with colors and visual indicators similar to those in code diffs, and warning and error segments, which allows marking a particular part of the code as having warnings or errors (with wavy underlines matching styles of popular IDEs).

linkRSS, SEO and Performance

As requested by the community, we also added RSS support to coding-blog-plugin. This means authors can get feeds in RSS2, Atom and JsonFeed formats automatically and with minimal configuration.

We also further polished automated SEO metadata generation of the plugin. In that light (but not directly related to SEO), we also introduced an early version of tags (via a custom tags component) that should help with further classification of content on the platform.

We did conduct investigations onto performance of coding.blog, and while the results are pretty good considering zero optimization done beforehand, there is definitely room for improvement (mostly to ensure that smart phones with bad connections will also have a snappy loading experience). Part of that effort is iterative improvements to CODEDOC and coding-blog-plugin, which we have already started with. Another part is even further improving our underlying rendering and state-management libraries, to trim them down to absolute basics while improving their speed. We want to sync this process with us polishing and releasing these libraries for public use, which we have started with the newest version of our client-side state management library, RxDeep.

linkThe Path Forward

In coming months, we will focus more on increasing the flow of content on the platform while keep improving the authoring and publishing toolchain powering it. As mentioned above, this increase will allow us to move forward the planned beta stage of coding.blog.

In that light, we are actively looking for enthusiastic writers. If you want to write articles published on coding.blog, simply drop me an email on eugene@connect-platform.com. If you are not enlisted in our prospective creators list, please enlist beforehand.

linkFuture Updates

We will keep you updated as we progress forward on this path. coding.blog is heavily reliant on the community to work-out, so we think extreme transparency in whatever we are doing is essential in its success.

You can follow us on twitter for more frequent updates and a regular does of new articles published on the platform, or stay tuned for the next update!