Hi Sara,

You need to decide if this is a demo app or an app which you'll want to scale and maintain. For a quick demo, I would spin up a Node.js server with server rendered frontend using Handlebars template. This way, the whole project can be maintained in one repository without much moving parts.

If you are extracting data from image itself and not from a database, you wouldn't even need Node.js. Just a React app will do. Data extraction from image can be done on client side.

Let's start with the database. First, in my experience, there are few applications where choosing a document database (NoSQL) over a relational database (SQL) is advantageous. While document databases are conceptually very straight forward, I find the tradeoffs down the road are simply not worth it (I wont get into all the details here, but please do some research on the downsides of NoSQL databases). If your data storage needs were exceedingly simple, I might reach for something from the Google Firebase suite, Realtime Database or Cloud Firestore; but I find even simple storage needs tend to expand and grow over time as your application matures. Postgresql is an excellent choice, and an absolute powerhouse for a ton of applications. With the somewhat recent additions of hstore, json, and jsonb datatypes, the advantages of reaching for a pure document datastore melt away.

For the Chrome extension, I would probably favour going for something a bit more lightweight than React or Angular. I'm a huge fan of React, but it comes with a somewhat hefty download, so if it were me, I might reach for Vue instead on that one. React is better for bigger, more complex single-page applications, whereas Vue is probably a better fit for simpler applications which require a smaller set of components.

For the backend, I would pick something mature with a strong and active community. Flask is a nice choice, but I've felt a bit "on my own" when using it in terms of community/documentation. I've used Rails extensively, but the learning curve is a bit of a headache; the time you'll save using Rails is very much down-the-road rather than immediate. If you're comfortable already with Javascript, then node + express is probably your best bet.

But, let me change my tune a little bit. You mentioned that this is a school project. In light of that fact I suggest you gravitate towards languages and frameworks that will help grow your career. Making smart choices based on the requirements of the task at hand is always prudent, but in this case I think it may be more valuable to gain some experience with some of the current "industry standard" stacks. Ask yourself what you can build a career on, and dabble in some of those areas until you find something that clicks for you. So, here are my revised answers, with options for each category ranked in order of preference

• Database: PostgreSQL, MySQL
• Backend: Node/Express, Rails, Django, Spring
• Frontend: React, Angular, Vue

Hi Karin, I really liked your take on this whole school thing, I'm amazed you want to put such a huge effort in it.

And please appreciate your project is a lot to take and it can also be a lot to do: the risk is going beyond the assignment for the sake of exploring technologies, architecture styles, desing patterns, and so on, just for the sake of it (don't take me wrong, I've done it all my life).

So my first advice, as quite an experienced software developer, is always go back to some fundamental principles before starting anything, before thinking to anything, and perhaps the most important principle of all is KISS: Keep It Simple and Short (search it up, there are a few versions of what those letters represent :) ). In your case, since it's a school assignment, simplicity is even more important because it makes things clear which makes learning so much more effective.

When dealing with complex tasks like this, another fundamental element is focus: where should you keep your attention when designing and then developing a software product?

In this specific case, I lack what the original assignment was requesting, but I'm quite sure the point (or one of the points) was to make you think and then act on something that didn't require months to be developed, it was to make you learn how to accomplish a task without getting lost in details or in a project too big to be finished in a finite time.

I may be wrong, but I'll keep this in mind when writing the below lines.

FIrst, the architecture of the software product looks like a classic three tiered one: frontend, backend, database. Keep in mind another fundamental principle here: the separation of concerns, which leads to different decoupled architectural elements. Also, just for clarity, the frontend(s) will talk only with the backend, while the backend will talk with the database: this will help you isolate the database from the frontend, ideally enabling you to change database technology if needed.

Second, you explained you want to go web and mobile for the frontend tier: this inevitably will lead you to the conclusions you pointed out correctly, having to choose a number of platforms and languages to basically create the same application, but the fragmentation of different knowledge and procedures can make your life quite complicated and probably miserable.

Personally I'd go for native Android and React for web. Recently, though, I stumbled upon Flutter which, through the same codebase (in Dart, very similar to JavaScript) can create for you applications for mobile (Android and iOS) and the web: I tried it and I've been blown away by the effectiveness and easiness of using it.

For the backend, keep playing with Flask and build a RESTful API, all in all python is a language way more readable and maintainable than JavaScript, and with node.js is so easy to fall into the callback hell (recently less and less but still). Stay away from Java and its ecosystem if you want to finish you project at all (just kidding).

On the database tier, remember NoSQL databases can be quite powerful, but in your case try something very simple (redis can do), or just go with MongoDB as it makes easy to start and evolve your data structures. If you're more the structured type and you want to go RDBMS, try postgresql, it's easy to start (it has also NoSQL features) but so much more powerful and you could learn real SQL on it (stay away from the omnipresent MySQL, it's kind of odd sometimes).

I hope the above didn't sound too much of a lecture, and I also really hope you learn the most important lesson of all: always keep in mind the big picture!

My Recommendations: Front End: Flutter because of developer tooling and powerful declarative widget system Back End: Node.js or Go because Node.js has a large ecosystem and Go has a good built in http setup Database: Cloud Firestore because of ease of use, NoSQL, and the ability to set data from the client

If you go with react / react native I advice you to go with node. Why ? I first didn't believe coding in javaScript everywhere (back, front and db queries) was making life SO much more easy. I still followed the advice, in the end this is a huge relief. For a small startup project with 1/2/3 devs, using only one langage increases efficiency a lot. You can switch very fast from a topic to another.

Considering that your objectives are Scalability, Reliability, Cost, and Efficiency, I recommend the following:

1. Backend - Node.js/Express/MongoDB
2. Frontend - React
3. Mobile - React Native

You can then focus on using one language (JavaScript) for all the above. As your startup grows, sure, you may have other requirements that warrant using other technologies. Nevertheless, the above stack will definitely satisfy your objectives. Hope this helps!

Flutter is a new sparc out there, because it's Dart engine can run server-side, client side (as web app) and natively - it cross compiles to all major platforms from single codebase...

I'm currently working with React and I would recommend you because it will help you develop both web app and similar to React is Reactive Native which will work mobile devices. And with these frameworks, I will choose Node.js in the backend. In DB, I have experiences with MySql and MongoDB. I think you should go with MongoDB, it will help out with its cloud service also. Happy Coding!😀🤩

For UI:- Flutter / ReactJS For Backend:- NestJs For Database:- Cloud Firestore & Mysql

We thought about creating a web application for a long time, but came to the conclusion that it is better to create an adaptive site with PWA technology. This will save your budget and speed up updates (you won't need to update 3 versions of apps for different platforms, just the site). In addition, research on the preferences of smartphone users suggests that users are not very willing to install new offers for reasons of personal data security. Sites that work through the browser are more trusted.

One thing I will advise you to is stick to what you know don't try something new just because it's trending or someone says it's faster and better unless you have the time to learn it. Just find a good folder structure that suits you.

This is what I use and why I use for a full-stack js application - Next.js ( for serverside rendering ) - Eslint and Prettier ( auto code formatting ) - Fastify ( fastest rest API in node )

I know how difficult is to maintain code quality and test coverage as startup, but keep this as a culture/rule to follow, because when things getting bigger, it's more difficult to keep flowing and evolve. More challenges and more technical debts will appear and stagnate the project.

I've heard good things about nestjs, but can't say much about because I've never used it.

I am using this stack and I think it's solid. I use Koa instead of Express which I find a bit more intuitive but the are almost equal in capability. I am using a bit of Typescript which I generate from the database (I've open sourced that: https://github.com/kristiandupont/kanel), but other than that I am not using a lot of libraries or frameworks. My current structure is a mono repo with Yarn workspaces. I have a client and a server folder, as well as a couple of packages that I haven't yet extracted enough to put on NPM. I use Webpack through Create React App for the client and also for the server because I have a few things that need transpiling. I use Gitlab for version control as well as CI/CD. Heroku for hosting and ElephantSQL for database. All in all this runs quite smoothly.

I suggest you to go with MERN Stack (Mongo,express,react,Node). As you know python and django which is a plus point because you can use python and node as your backend and for front-end use react(easy to learn) and database of your choice.(Mongo or SQL)

GO For MERN Stack... brother

You are probably referring to ruby on rails for web development and nodejs for building the backend. Nodejs has frameworks such as express and next which not only provides a minimal code to build a backend but also gives the flexibility to try and experiment with the framework choices. For example you can have express framework + Passport for OAuth .... etc. The flexibility and the constant improvement of the language provides a good reason to opt for nodejs. Nodejs uses javascript which makes your code uniform when you are working full stack i.e react in front end and nodejs in backend.

I'd use the following metaphor to non-technical people. Rails is like a prepackaged toolkit, which can get most of the common tasks done fairly with ease. Whereas, node.js with whatever backend framekwork of choice, is like a DIY toolkit assembled by mix-and-match different tools in a large tool shop. Of course, at times DIY toolkit can do better on specific tasks. Given that you are the only dev on the project, I'd assume that the resource is fairly limited. And looks like you are not building some next-gen super duper fast smart application. So Just go with the prepackaged toolkit then. Rails is a very opinionated framework, there're pros and cons to it. But thanks to that, many of the gems are coded with it in mind. For example, they are all designed with same naming convention. Many will work well together out-of-box, for example devise and cancancan. Besides, many stuff are built in the framework. For example, logging utility, csrf protection, session encryption, etc. Yes, many of those stuff may not be useful or necessary at the beginning of the project life-cycle. However, down the road, there is a good chance you will need some of those. And the moment you realize that you already have it, it's so delightful. In addition, it's usually easier to debug a rails app than a node app in my experience. Personally, the cases where I would pick node.js over rails would be projects either require a) high-performance, or b) certain core functionality that has been implemented by some node packages but not by any ruby gems. In term of performance, node has a clear advantage over any other major web frameworks, except the ones built with go. It's simply a language feature. Node allows developer to easily write code that runs db query, external api calls, or other stuff of that nature in parallel. And that is THE MOST COMMON performance bottleneck of web applications.

Rails is currently a very mature and feature complete framework.

It's the ideal one if you're the only dev for your project because you get so many things already baked-in the framework that you'd only need to deeply care about specific stuff.

I won't say any NodeJS framework isn't good enough but in my experience with NodeJS frameworks you have to code a lot of the things Rails already provides. There's many people in Twitter and IRL asking for a "Rails for JavaScript" framework.

And you know? In the early stages of any project we have to validate it first with real users/customers. With Rails you can get to production real quick and fast.

I'm going to mention some of the features you get from day 1 when you run rails new app_name:

• File uploading with Active Storage
• Rich text editor with Action Text
• Emailing with Action Mailer
• ORM, migrations, validations with Active Record
• Web sockets with Action Cable
• Internationalization
• Modern frontend stuff with Webpacker

and more.

The JavaScript community is on its moment, growing and gathering more people everyday but the Rails community is also a big one and there's always going to be a Rails developer to hire whenever you're ready to hire someone.

I suggest you to go with Rails because is a good choice, gives you less things to worry about and it's a very good and mature framework.

I hate to admit it, because I loved my time with Rails (and I still love the framework), I have a hard time justifying new Rails applications these days. Core team has made some tragic design decisions, and developers just don't perceive it as being "cool" any more. The latter is a terrible metric for which to base a technology decision, but I think you'll find it more difficult to recruit additional engineers if you choose Ruby on Rails.

Without knowing too much of the details, Node/Express (ideally with Typescript) seems like a better solution here, given you'll be building out the front-end in Vue or React. It might be worth looking at NestJS, as it's the closest I've seen to a well-formed opinionated framework on the Node side of things. We're also fans of Objection ORM.

I hope that's helpful!

I need a solid argument to explain to people that aren't necessarily tech-savvy as to why we should choose Rails over Nodejs

Hi Max, it sounds like that you are proficient in both stacks and probably have a higher expertise in Rails (correct me if I am wrong) and since you are the only dev on a project, a good argument that comes to mind is probably the velocity and maturity (enterprise grade, battle tested in production) that Rails provide with proven success stories in the tech industry such as Airbnb, Stripes, Shopify to name a few. You can also make the argument that Rails is great to run the backend and React+Vue (and nodejs for tooling) is ideal for the front-end development (see or find companies example that use both). You can also build and show a prototype using both and share your experience which could help you find and forge the selling points to those non tech savvy folks, why not.

Eventually, are you going to have other developers on your project? if yes then you will need to take in account, onboarding and ramp up to contribution time when they are hired.

IMHO, I am not a fan of the debate Rails vs Nodejs, they are just tools at the disposal of the developer it's just a matter of figuring out what makes the most sense.

Let me know if you wanna discuss further, happy to help out!

ps: markdown preview on stack share... no good.

Rails has advantages over node.js (specifically express) when working a more complicated backend. While Express has some speed advantages to Rails, this is mitigated if your software is more CPU intensive.

If you are currently not working my first suggestion is to study both the frameworks and get a good grasp of those. If you didn't get confident with Django in the first place you should reconsider going back and study more. Get a video course with some code-along and produce some simple application you can showcase on your interviews. If you already took a course take a different one. Another trainer could be more effective and you could experience something new with different excercises. There are lots of both free and paid courses out there. When you will get confident with Django get your feet wet with Node.js because it surely worth it. Node is very different from Django from some perspective, it looks more like an asynchronous version of Flask to me. Be sure to have a good knowledge of ES6 first, because it will be really useful to understand the Node best practices. Study as much as you can now if you are not working. It will supercharge you for the future...

From my experience of the early startup world, a majority of companies these days use Node.js. Python and Go are the next biggest languages, but significantly smaller than Node.

However, if you're having trouble with the front end aspect of Django, using Node probably won't make that easier for you. You'll have a lot more options between front end frameworks (React, Vue.js, Angular 2) , but they'll definitely take more time to learn than Django's templating system.

Think about whether you want to focus on front end or back end for now, and make a decision from there.

Jinja is a template rendering engine and you will encounter some sort of template rendering engine in each language. Jinja is a pretty standard tool and almost every language has some sort of Jinja equivalent. Ruby has Liquid, Node has Nunjucks, Java has Jinjava, Go's default templating engine is easy to pick up if you know Jinja, Helm charts are easier to pick if know Jinja . So learning Jinja is a good thing.

I had the same question myself a few months ago. I finally chose Node.js, and it was one of the best options I did back then. From when I started programming, I always believed that Python was for me the best language, secure and stable. However, it is not flexible for web development, there are more packages & libraries that are built and work only with JavaScript / TypeScript, and the community, resources & support is much bigger. I was also fascinated by the Django ORM, which I still am, & the admin interface. But those are things, that can be replaces by other tools, such as TypeORM, and the admin interface was not needed at all finally for my case. I know understand that Python is not the language that I should use everywhere and every time, but I can say that it is really good for algorithms, computer science, maths, statistics, analytics & AI. To be honest, I chose TypeScript (TS) with Node.js & Express, because it has auto-completion and "strict" code checking. I hope this helps you, and let you take a look at various aspects of choosing a programming language to work with.

I would suggest to go with js, it's the craze now when you enter into the stack it has variety of options and tools that you can adopt , and more than that the demand for js engineers is exponentially increasing and js can do magic in any type of application or architecture.

If you already know some django stuff you should keep that learning path. And for the job if you really want an internship you should learn to make rest APIs using django or nodejs, and a front end that consumes those APIs using some framework

Actually, you could get very good solution with implementing BE and admin panel with Django and FE with React.js or Vue.js. it will provide you a pretty flexible and powerful environment.

I highly recommend NestJS because:

• It's a framework you already like;
• Typescript is growing fast, being increasingly adopted in the community;
• All layers are well defined, not needing to think much about the organization;
• Great documentation;
• Nest CLI increases the development speed and keep the pattern;

Only using express and knowing that project can grow, you'ill need to define the structure well so that it doesn't get out of control.

Hi Louai,

I am quite sure that you know the answer to your problem. And I am here to help you to follow your arrow. I have worked with the most popular Nodejs frameworks and I can sure you that there's no stack better than NestJS (at all). Typescript is the best thing that happened with Javascript, this is a fact. Ans NestJS make a such wonderful job using all the best Typescript tools. NestJS is the most mature and organized API manager. Its modular dependence injection, the use of DDD, the solid idea of single responsibility, it's unit a and e2e testing support, its documentation is the most incredible work in the world of Nodejs. You won't regret choosing this framework, even if your application grows a lot. If you follow the documentation tips you will be able to create an amazing and organized application.

ps: I am not part of the NestJS team, I am just a guy tired of wasting time with dumb and bad Frameworks and its bad documentations. I find relief in NestJS with all the time it's saved to me, it helped me to improve my job and let me create great things with Nodejs.

First of all, my experience using either Node.js with Express or NestJS is not wide. I liked NestJS due to it's similarity to Angular, so when you know Angluar and like TypeScript you are going to love NestJS, it will be instantly very familiar and easy to use, it's adds a good structure to the project out of the box and well, it uses TypeScript, which is a more structured language - it's good for scalability. As for performance concern s - NestJS is based on Node, it just brings Angular's modular structure to it, so the question is more about how is the additional layer influences the performance - I cannot answer that.

Hi, I'm in a similar position, but related to personal projects. After falling in love with few frameworks in the first day and rejecting them in day 2, I started learning nestJS last week. I currently develop personal side projects using cakephp, and I intend to migrate to nest + vue. This week I'm taking a nestJS course in order to be sure that this is what I want by praticing a little. If you didn't do it yet, I suggest you try to code a todo app or a similar example API using nest, so you can "feel" if this is indeed what you want to use in this larger-scale project.

Some of the characteristics that got my attention to nestJS are typescript, a lot of annotations/decorations, an oppinionated approach to organizing the project, nice documentation and discord, and it's evolution at npm trends shows me it's probably not going to vanish or get buggy anytime soon.

I would definitely suggest NestJs over other options because NestJs gives a lot of tooling. it would definitely suggest NestJs over other options because NestJs gives a lot of tooling & it gives a lot of functionality out of the box. If your team worked with angular 2+ then it will really easy to learn.

Have you checked out Hapi as an alternative? I'ts not Typescript by default though. If that doesn't seem too interesting, it sounds like you want to go with NestJS :)

I recommand you use the mean stack. This is really effective :) MongoDB -> database Express -> backend api Angular -> frontend NodeJs -> backend server

You can use react instead of angular and firebase instead of mongo. This is a good place to start :)

It's probably worth investing some time for your team to learn Node.js. There's very little overhead, especially with a framework like ExpressJS, so if your team is familiar with JavaScript it should be a quick process.

It handles I/O really well out of the box, and has a strong community with great open source libraries. Since you're using React on the front end, there's also some benefit to being able to use JavaScript throughout your stack.