Mo Morsi

Having worked on projects of all scopes and sizes, I suggest not overthinking the refactoring process. Ideally it wouldn't be needed, with experience, quality software can often be written the right way the first time, but sometimes you can't get around it. In these situations the team will almost always know which areas of the codebase need love ahead of time as well as the design patterns best suited to shore them up. Tackle things incrementally (tests are your friend here so as to not introduce regressions) making sure to undertake a methodical & systematic approach to ensure the project is continuously moving forward (again experience is gold here). At the end you will have a well designed product that meets customer needs!

If you have a choice to pick the tech (as opposed to having it prescribed to you by an employer or professor), you can't make the wrong choice. Sure some languages are more conducive to developing skills which are in demand (if that's your goal do research pertaining to what's being used by the industry you're targeting) but even esoteric or obsolete languages can provide very valuable learning experiences. The most important thing when it comes to any technical craft is discipline & repetition, particularly around learning the fundamentals. Regardless of the language, it's the same x86 or ARM architecture under the hood (unless you're doing something specialized) & best practices learned in one language often transfer to others.

This is a tough question due to in part the many scenarios which have disparate solutions. To assist those with visual impairment then better contrast & audio prompts go a long way. For individuals without access to high speed Internet, then optimizing resources & load times greatly serves their needs, etc While a noble goal both from a humanitarian and a business perspective, the ability to execute here is highly dependent on the resources at hand. Unfortunately many small businesses can't afford to improve the experience for *everyone* so talking to users (both current & potential) will help address any issues that they have. As the product succeeds, more resources can be dedicated to building out a service that is accessible to all.

This answer really applies to any skill in any domain. The key here is discipline, the more you "do it", the faster you will improve. Some times it will be frustrating and you will feel as if you are spinning your wheels but as the months and years go by you'll notice a gradual improvement to the point which you are able to consistently produce solid software on the first attempt, no matter the language/framework/environment/etc