Former Microsoft engineer: Originally Windows was developed like this

Computer store news: Two days ago, a former Microsoft designer talked about how the company decided to use the burger menu in Windows Phone to dramatically replace the pivotal UI. The designer said that the decision was supported by a large amount of data and was also the result of a joint discussion by the design team, and many very smart team members agreed with the decision.

Now, there is another piece of content about the Windows development process shared by Windows engineers more than a decade ago at Microsoft. These contents show us how a good idea is Destruction, and how a bad idea is ultimately adopted.

He wrote:

I used to be an engineer on the Windows team and participated in many development seminars.

Often, some extroverted and charismatic project managers will announce that the team has made a decision. However, this decision is usually a product of multiple compromises, which may be functional, or applicability, or time constraints. The original ideas are often very good: fast, widely adapted, intuitive, automatic, and most importantly user friendly. At this stage, the conception phase of the next-generation Windows operating system, it is often seen as a very great product.

Often these designs are reviewed by users and partners with functional approvals. The product at this stage, the next generation Windows system, is a great product, please believe me. In the product release process, a series of decisions continued to reduce the functionality of Windows, so that when it was officially released, the final product was very different from the original product.

The problem is that all of these decisions don't look so bad throughout the process. Because the project management team, developers, and testers have been carefully reviewed and have spent a lot of time weighing each choice they made, assessing the pros and cons of each option and estimating its possible impact, ultimately Choose from a number of choices that you think are the best.

Then the question is coming: Who is not involved in the decision-making meeting? At this stage, usually the time period is 4 to 6 weeks, there is no time to consult the user.

In theory, almost all members of the Microsoft engineering team are considered representatives of users. Project managers have already understood the user's ideas when planning, so these features can predict each of their needs. The tester first noticed the user's dissatisfaction. Partners are somewhat like users, but they are more money and more influential. As for developers, they don't actually care about users.

In fact, this process does not work. Because no one in the Windows engineering team really touched the user. They are actually passing on the idealized image of a general-purpose user.

The functional team consulted on this divination-style model, and then carefully rationalized all the ideas in the air, and then became a decision. But this decision is ultimately based on the engineering team's ideas and ideas, not the user's ideas, simply because the users did not have the opportunity to attend these meetings.

Maintaining multiple user interfaces for a product is an extra task, and we don't have enough time to test multiple configurations. Other competitors have more market share than us, and if we just copy them, maybe we can do better. All of what we say is true, and all of this is a strong reason to support the final decision. But all these reasons and arguments have been answered by Microsoft.

On the day of the meeting, project managers will repeatedly emphasize how well the decisions they make are beneficial to users. They will cite some facts to support this decision: we don't want to provide too many options to make users unable to choose; consistency is good for Microsoft, so it must be beneficial to users. Everyone with a satisfied smile, nodded and agreed to this decision and agreed that this is the best decision.

Although the final result is somewhat different from the original idea, the user experience may be worse, but writing software requires compromise. This is an acceptable compromise, and it is not too bad in any case, it is also the best choice. If there are users on the scene, they will understand.

The above is what I saw when I was a Windows engineer.

The former Microsoft engineer’s implication is that although the data and decisions made by Microsoft are not always correct, if no one inside Microsoft truly represents the user, then when there is opportunity to feedback, Of course we should not be silent again. At present, during the development of Win10, Microsoft has made the Windows development process more transparent through the Windows Insider program, and began to listen carefully to user feedback. So, everyone is actively sending out your voice to Microsoft!