Jess's Site

It's been almost five years since I last posted an actual new post on this website, with any changes mostly being minor content fixes rather than brand new posts. However, to coincide with some personal news I have posted on Twitter, I figured I would post something new to summarise the last 5 years, with some more in depth posts possibly coming sometime in the future. Without further ado, here is a summary of projects: - My final year project was a software system designed to be able to manage multiple CCTV cameras with differing recording schedules without a central server, instead any device on the network could connect and control the cameras via mDNS using a Windows app. - I did a long form post with my employer about my experiences with making high performance React Native applications on both native and web, and how we had to move to Tamagui due to poor performance on the web. - Made a Node.JS payload app to create AI powered chatbots you can add to a Discord server - Various other AI experiments I want to clean up and open source at some point. I also took the opportunity to rebuild my website once again: It's now running the latest Payload 3 as a CMS platform with Next JS 15 for the frontend technology.

So firstly -- WinIRC is not dead. I am still working on it, even if there's been no commits to the repo, I've still been working on fixes and new features in IrcClientCore, which will end up in WinIRC. Eventually, I hope to finish the current update for WinIRC, which although isn't the recode I mentioned in the last release, is a major decoupling of the IRC logic from WinIRC into IrcClientCore. It also has a number of UI tweaks including coloured nick names, and displaying the number of unread messages in the channels on the sidebar. This update is planned to happen soon. The reason I've not been working on it as much (apart from university / being employed) is a WIP project currently called Signal IRC. This is an IRC client based upon the same IrcClientCore library as WinIR C, but rather than running as a native UWP application, it's an ASP.NET web application. This is essentially the spiritual successor to an older project I had, namely Comet IRC. The major differences is that the backend and frontend is not nearly as tightly coupled -- the backend utilises a mixture of ASP.NET Web APIs for login / server management and SignalR for real time messaging. The frontend is a Vue single page application that's served by the ASP.NET server portion. I plan to have a public release of Signal IRC as an alpha on Github at some point this year assuming I have time, once a few key features are completed, including storing message scrollback within a database.

WordPress has served me well for a good few years now, but I decided last year that it's time for an overhaul of my site to use more modern web development practices and for me to have much more control over it. I also wanted much more emphasis on the projects I've done, rather than have them buried as a small link on a subsection of the site. This led me to create this site -- a Hugo powered static site with a couple extra server side scripts to manage Netlify style continuous integration and GitHub auth for the Netlify CMS used for content management. It also utilises the InstantClick JS library for Gatsby like SPA navigation and page preloading in a couple KBs of JavaScript The site is fully open source, and available to download at https://github.com/rymate1234/rymate.co.uk.

WinIRC 1.5 is finally making it's way to the Windows Store, with a host of new features - the main one being a refreshed UI. However, the release of the application has left much to be desired. Just over two weeks ago, I pushed the final 1.5 changelog to the WinIRC Github, as I was hoping in the next couple of days the release would be live on the Windows Store. As it turns out, this wasn't the case - I forgot to test it locally, and the Store based certification process marked two issues that needed to be fixed before it would be certified, specifically: - The twitter library used had uploaded binaries to NuGet built in debug mode - Metadata issues with the UWP MenuBar lib I created So I fixed the first issue by building a local NuGet package for the twitter library, then tested with the local Windows App Certification Kit (WACK) and my app passed, with no mention of the other metadata issues. So logically, I tried to release it on the store, and it failed the store WACK test with the same errors as last time! I first assumed it was my machine, so I tried it on my tablet and then a Creators Update VM, and failed to reproduce on both After trying (and failing) to get help from MS support, I tried last night to rename the filename of the package that was uploaded to the store, as all of the ones I'd uploaded were the default filename. This worked! Although there was still a difference with the store results to my results, I managed to fix the metadata issues with my UWP MenuBar (by importing the source as a c# class library to the main WinIRC source tree) meaning WinIRC passed it's validation tests. So what's next? For WinIRC development, the next step is to refactor the app to properly use an actual design pattern, probably MVVM. I also want to add some scripting support to WinIRC, so users can add their own commands and functionality to WinIRC. For distribution, I want to work on alternative distribution channels for WinIRC and other UWP / AppX applications, so it isn't entirely centralised in the Microsoft ecosystem, and hopefully so apps don't need to have a special signed certificate. More on this if I actually manage to make a prototype

A couple months ago, I posted on Twitter that I'd been working on a menu bar to be used for UWP applications: To mark a sort of return to twitter - a screenshot of something I've been working on pic.twitter.com/R5jYX5wyFX— Ryan Murthick (@rymate1234) March 13, 2017 A couple days ago, I released the lib on NuGet and GitHub as a prerelease! Here's a video demonstration of the library: https://rymate.co.uk/videos/uploads/670f40e323.mp4 Head to the GitHub for a detailed readme and usage docs

So today I decided to make a lightweight theme for this website. It’s based upon the underscores.me starter theme in order to make things easier for me. Although it’s pretty barebones at the moment, I’ll be making it look a bit snazzier over time. For instance, I want to put the site sidebar in a hamburger menu when on mobile devices and other smaller screens. I was going to give it a style similar to Google’s Material Design, then I realised something like that would be absolutely pointless for a blog, so I decided that instead it would just be a reasonably lightweight theme. Also the blog post on setting up a print server on a Pi is definitely coming soon probably never happening.

It’s been over a year since my last post – mainly due to the stress of AS levels, and then settling in to a new college. I have at least kept the website up though! I’m going to start posting more stuff on here – now I’m doing a ICT and Business course at college, there’s some small projects I’m making  in the ICT part of my course as part of my coursework that I may put up here.  There are, of course, be other projects unrelated to my coursework – I may post about them here! One such project with my Raspberry Pi is that I made a print server with it that allows printing across the network with a samba server for windows devices and cups printing for all other devices, which I will post about in the next week or so. I’m also working on numerous improvements to Rymate Notes which may or may not surface.

This is the largest release of my app ever, and thus I’m actually bothering to blog about it. This release adds a lot of new features, such as formatting notes and a quick edit feature, for when you spot that misspelling in a long note! Also in this release, I removed support for older devices still on Gingerbread. This was mainly to make development of the app easier, as supporting legacy devices such as gingerbread wasn’t worth the extra effort. DOWNLOAD HERE

So I’ve been using my iPad for a few months now, and although it is a great machine, it has some bad points. Here is my rundown of the iPad mini. Good points: 1. The screen is simply the best screen I’ve used on a mobile device. It’s the clearest screen, and it is very good with watching youtube or the BBC iPlayer on it. It might not be a Retina display, but to be honest I’ve never really noticed the pixels when using it. It’s also a really good size – After using the iPad mini, the standard iPads just seem too big now! 2. The UI is silky smooth. When using apps, I very rarely experience any form of UI lag. Scrolling is smooth, and there appears to be none of the jerkiness that I get with my android phone. The only UI lag I’ve seen is in Bad Piggies when I make some massive huge vehicle in the sandbox, and oddly enough Skitch and Evernote is really quite laggy on occasions. 3. iPad apps are really well made. I haven’t found a single bad app yet, all of the apps seem to be well designed and easy to use. This might be a huge advantage of the iOS walled garden that seems to be a major complaint about iOS. Sure, you can’t install apps from 3rd party sources very easily, and apps on the app store might have to go through a rigorous process to get onto it, but it does mean that the vast majority of the 800,000 apps on the App Store have a very good user experience. Bad Points 1. The typing experience is terrible. I sincerely hate the typing experience on the iPad mini. It isn’t terrible, but there’s quite a few annoyances that I really hate. I quite often type quickly on the screen, and as I have big hands it means I quite often make typo’s, one of the most annoying ones being when I hit the “m” or “n” key when I want to hit the spacebar. When it does happen, more often than not I do it again so autocorrect doesn’t auto correct it. Selecting text on the iPad is a pain too. When I hold down on the screen to select text, a magnifying glass shows up to help you select the text. More often than not though, the iPad “cleverly” attempts to highlight the entire paragraph when I only wanted a single line. 2. iOS is severely limited. I can’t change any of the “default applications” that Apple forces upon us. I might not want to use safari as my browser, instead I want to use firefox. But Apple thinks all iOS users need is safari, and that alternate browsers are evil. You can’t even download firefox on iOS, and all other alternate browser are either a webkit wrapper or a “cloud based” browser.I can’t change parts of the UI either. On android, if my phone came with a crappy keyboard, I’d just install another one off the google play store. I’m stuck with the iOS keyboard. I want to be able to change aspects of my device. I like the idea of Apple’s walled garden, but it’s like a garden which you can get some excellent flowers for, but you can’t change the tree in the middle even though it looks ugly! How apple could fix the bad points All apple needs to do to fix the bad points is make the iPad more like a computer. By that, I mean allowing users to change the system keyboard, or change the web browser, or choose what App they want to open stuff in. If Apple doesn’t want users doing that, make it a setting that allows users to turn on “Advanced Customisation” so the average joe can’t just install new keyboards and then not know how to use them. Other than the bad points mentioned, I think the iPad is a really good device, and I would certainly recommend it to people who already have an Apple product.

I’ve moved the website to a new webhost that should be a lot more stable than the old one 😀