Journal
Journal: comparisons
On this topic, 40 articles from the journal, from practical guides to behind the scenes of the product.
FlutterFlow or Bubble in 2026: one exports code from $39, the other keeps everything
FlutterFlow or Bubble: the comparison depends little on features and heavily on the exit. One lets you download application code from its first paid tier. The other's documentation says leaving is impossible.
Read the article
Freelance developer day rates in 2026: the surveys and what a day buys
Two source-accessible surveys give median freelance developer rates between €454 and €520 daily in 2026, with a €580 upper quartile. What the numbers cover, what they hide, and what a day buys in practice.
Read the article
Cursor vs Windsurf: the latter is now Devin Desktop, and the question has changed
Searching 'Cursor vs Windsurf' in 2026 leads to an outdated comparison: Windsurf changed names through an automatic update in June 2026, and its old pricing page redirects. What remains to compare, and what neither does for you.
Read the article
Emergent review 2026: the autonomous agent targeting France, and the autonomy argument
Emergent builds an entire application while you watch, claims half a million applications created in France, and sells autonomy as its advantage. What it does better than us, and the three missing pages.
Read the article
Building an application with ChatGPT: what you get and what is missing
Building an application with ChatGPT works up to an identifiable point. What the conversation provides, what OpenAI's coding agent adds, what it costs, and the three things it will not do for you.
Read the article
Airtable and Notion as applications: the limits you discover in month six
A shared database solves in an afternoon what a software project would take weeks to build. The wall comes later, with an access rule, external guest, or bill line that has changed nature. Here are the limits as they appear.
Read the article
Our Manus review in 2026: the autonomous agent that does everything except let you decide
A Manus review depends on what you ask it to do. For producing market research or a report overnight, the autonomous agent is the market's most versatile. For building software you will maintain for three years, the question changes.
Read the article
Build a mobile application with AI without coding: what remains before publishing
Three tools now build mobile applications from written instructions, and one submits to the App Store in two clicks. What they charge, what you take when leaving, and the part of the work none of them writes.
Read the article
Codex or Claude Code: the choice explained for someone who will never open a terminal
Two coding agents with the same entry price, two ways of counting consumption, and a French question with no answer online. What each does well, what neither requires, and why the choice should remain a setting.
Read the article
No-code agency or freelancer: two ways to delegate, and what each leaves you
A no-code agency quotes €5,000–€15,000 for a first tool; a freelancer's median rate is €520 per day. The figures compare poorly because they buy different things. What separates them appears in month six, when you want to change a line.
Read the article
v0 or Lovable: one builds a screen, the other an application, neither writes the specification
v0 excels at interfaces and fits into teams that already have review rules. Lovable puts a complete application online in an evening. Neither produces the document stating what the product must do, and that absence has a later cost.
Read the article
Softr or Glide: two internal tools, two ways to count users
Softr charges for builder seats and end users; Glide charges credits whose intensity you set yourself. Both turn a database into an internal application. The choice comes down to three lines: who counts, where the data lives, and what you take when you leave.
Read the article
OpenAI DevDay 2026: what non-developers should watch, and what they can skip
OpenAI holds its DevDay tonight in San Francisco. This article appears before the keynote, so it reports none of what will be said. It offers something else: a way for a business owner having software built to interpret it, and a list of what to skip.
Read the article
Do you need French AI to develop an application? Mistral, Dust, and your data
Mistral has signed with Microsoft, opened European regional access, and is targeting a gigawatt of computing capacity. What remains is to understand what that sovereignty covers when AI writes your software, and where the boundary lies between what stays with you and what leaves.
Read the article
Bolt.new review 2026: a million free tokens, a bill depending on you
Bolt.new offers a million tokens monthly, then charges what its agents consume, without internal credits or opaque conversion. The trade-off: your bill depends on how you ask, and nothing requires Plan mode.
Read the article
Choosing an AI assistant and model in Maestro
Understand assistants, models and agents, compare two options on a small task, and prepare a switch without losing your project decisions.
Read the article
Build an application: choosing between local and cloud tools
Separate four places: construction, project files, AI processing and the finished application. Compare access, costs and recovery through a practical trial.
Read the article
Base44 review 2026: acquired by Wix, and European data at $160 monthly
Base44 takes over a Salesforce or WordPress application with real data, something none of its competitors can do. But it charges for European data hosting at its highest tier, and its exit opens only one way.
Read the articleModel-provider dependence: your AI tool may lose its engine on November 12
OpenAI will cut Cursor's direct model access on November 12 for reasons unrelated to Cursor users. A development tool depends on a supplier it does not control: here is how to measure that dependence before choosing.
Read the article
Per-user pricing punishes hiring: your small-business software with six people
Twenty-five dollars per user seems reasonable until the sixth hire. Then the higher tier becomes mandatory and the bill doubles. The mechanism, figures from business-owner forums and two ways out of the curve.
Read the article
Replit review 2026: the market’s best scoping, and what belongs to you
Replit does one important thing better than its competitors: it has you review a work plan and gives you a yes button before code exists. What you take when leaving remains the question, and Replit answers it poorly.
Read the article
Alternative to Bubble: the wall is an access rule, not a feature
Nobody leaves a no-code platform because a button is missing. They leave when a customer asks who may see which row and the tool has no place for that rule. How to test this wall before hitting it.
Read the article
Claude Code or Cursor: which should you choose without coding skills?
Claude Code and Cursor without coding skills: interfaces, prices, plans and checks. A practical comparison to help you choose, including where Maestro fits.
Read the article
How much does an MVP cost in 2026? Budgeting a first version
An MVP is a usable first version for testing your idea. Here is how to limit what you build and prepare a budget, with a fictional workshop-booking example.
Read the article
Bubble or Lovable in 2026: which should a non-coder choose?
Both tools can help you create an app. Focus on everyday changes, expenses and what you can retrieve if you switch solutions.
Read the article
Custom software or SaaS: comparing the cost over five years
The cheaper option depends on the features you need, the people using the software and the work involved. A fictional five-year example compares invoices, your team’s time and the cost of switching.
Read the article
Lovable pricing 2026: free plan, Pro and credit rollover
Understand the free plan, Pro pricing and credit rollover before paying. Then plan for the cost of running your application and check how to take its code with you.
Read the article
How long does it take to create an app in 2026?
A screen to show, a tool to try or an app used every day: the timeline depends on the result you need. Here is how to prepare a schedule you can follow.
Read the article
Can you keep your application if you change tools?
Your customers and their appointments are already in the application. Before depending on a tool, check what you can retrieve and what you would need to continue elsewhere.
Read the article
No-code or AI: is learning Bubble still worthwhile in 2026?
Unsure whether to use a visual editor or build an app with AI? Bubble offers both. The choice mainly depends on how you want to change and maintain your tool.
Read the article
How to choose job management software for a trade business
Keep, extend or replace your software using one realistic job: an accepted quote, a change, a phone photo and the return to the office. A sheet to compare tasks, costs and exports.
Read the article
Why we also describe failures in our AI agent tests
A good score can hide a missing feature. Two results from our internal trials show why we need to examine errors and specify what we actually measured.
Read the article
Can a team of AI agents deliver software? Our tests and their limits
Our tests covered an initial development batch. Here is what they showed, how we evaluated it and what still needed checking before real use.
Read the article
Why Maestro moved away from BMAD: our decision and test results
We replaced BMAD with our own method to organise Maestro’s steps more directly. A campaign of nine scenarios in August 2026 informed that decision without establishing a general ranking of tools.
Read the article
“Waterfall strikes back”: our response, point by point
225 points and 191 comments: the Hacker News thread accusing spec-driven development of reviving waterfall made an impression. Three precise criticisms, a quantified response, with no evasion.
Read the article
Lovable, Bolt, v0 or Maestro: which AI tool should you choose for your application?
Compare project preparation, construction, publishing and code handover. A reading of documented capabilities to help you choose for your needs, including Maestro's limitations.
Read the article
How much does an app cost in 2026? Planning your budget
Building the app is only part of the bill. A worked example and the expenses to compare for an agency, a freelancer or a project built with AI.
Read the article
Six days of testing: what we measured against BMAD
From August 5 to 10, 2026, we compared versions of our method with BMAD and then with each other. The results concern documents and an initial set of features, not complete applications delivered to customers.
Read the article
Agency, freelancer or AI: who should build your app?
You want an app, but you do not code. Here is how to choose based on the time you can give the project and the help you will need afterwards.
Read the article
What our ×10 measures: instruction text, not your bill
Our historical measurement compares the instruction volume of two methods. It does not mean ten times cheaper or ten times better. Here is the calculation and its scope.
Read the article