0 reply content: from another perspective:
One of the hot topics in the industry today is that Facebook founder and CEO Mark Zuckerberg admitted in an interview that "focusing on HTML 5 is the biggest mistake he has ever made .」 Then, the leaked data is that the user browsed the Feed information stream twice as much as before.
What are the main reasons why Facebook initially used HTML 5? For a single development, cross-platform release is definitely one of the considerations. Of course, development can achieve rapid iteration, which is also consistent with Facebook's engineering culture. However, this actually saves the development cost and achieves the "Speed" of development. This also sacrifices the "Speed" of user experience and the "performance 」.
Why is it wrong for Facebook to over-press HTML 5 on mobile? The biggest possible cause may be the performance issue. Without a better speed, there will be no better user experience, and user experience has always been the most important thing for Zuckerberg.
Since Facebook was founded, Zuckerberg realized that for network services like Facebook, "performance is the key. If the speed of transmitting new pages to users begins to slow down, it is a fatal blow .」 From a technical point of view, Facebook has always spared no effort in website optimization. Whether BigPipe or HipHop for PHP, these optimization practices and technological innovations have brought an excellent user experience to Facebook, mobile users' bet on HTML 5 is just a coincidence of Facebook's rule.
After the release of iOS native applications, what does browser information mean twice the original browsing information? Do users care about the local native applications you develop or use with HTML 5? Most users don't even care about this. Users are more concerned with "application speed". Is the App fast enough? Whether the information can be read more smoothly. No one wants to wait for an application to open slowly on the mobile phone.
The challenge facing Facebook is whether to establish more effective R & D mechanisms on mobile products like those best practices on Web products. After all, this is another battlefield, is an Internet giant absolutely dominant in the mobile field? No one knows.
It is not important for Zhuge Liang to evaluate the right and wrong of the incident. What is important is, can we learn some lessons from it?
-- EOF --
For HTML 5, there is no "blow". It may be a good thing. let more people know the advantages and disadvantages of HTML 5, rather than rushing forward.
I once said this at the time last year, "my two stubborn ideas: 1. HTML 5 is not the rescue star of mobile development, at least not now. 2. Because there is 1, therefore, solutions such as PhoneGap are not reliable and there is no silver bullet. It will take another 18 months .」
Now it looks like it will take another 18 months.
This article is first launched by Forbes Chinese website (URL ). Reference from http://internet.solidot.org/internet/12/09/12/0216205.shtml
Mark Zuckerberg indicates that putting it on HTML5 is Facebook's biggest mistake. Because
HTML5 applications suffer from poor performanceFacebook had to rewrite all HTML5-based mobile applications to native applications. He also said that this is not just a problem with mobile devices,
Over-believing that HTML5 is a problem.
From the perspective of the programming language trend, the decrease in javascript and the increase in objectiveC prove that HTML5 is not suitable for mobile application development,
No matter how GPU acceleration is used, HTML5 graphics rendering is still dependent on the CPU, which cannot meet the current demand for software smoothness..
Now it seems that Quora's combination of HTML flexibility and native UI of local applications may be better. Estimate the efficiency and development speed of HTML5, invest two years to develop and improve a set of frameworks, so that the development of multiple platforms can be integrated, the product changes can be accelerated to push new features faster. In the end, HTML5 is not ready for use, and the performance is limited. The user experience of mobile devices is not good, you cannot set the target to the reverse direction. Later, I switched to Native again. As a result, the two-year period of time also caused some possibilities. I think this has a lot to do with Joe Hewitt. He is basically a web platform man, and he has influenced some of Facebook's major decisions. The reason is that Zuckerberg discovered his stupidity.
HTML5 is superior, that is, it is no different from native. It sounds like a fart.
In addition, it is difficult to achieve the ultimate because of the large device differences.
Facebook does not actually have any advantages after HTML5, but has a lot of disadvantages. It is difficult for users to understand HTML to achieve this level because various compatibility problems reduce UE.
Linked in also gave up HTML5.
I want to talk about it again, but forget it. No one looks at my answer. One thing I think Facebook should reflect on through this incident is,
If you choose a technology as a platform and place a bet on it, this mentality is wrong..
For large technical companies like Facebook, including Huawei, the most worrying platform selection problem is"
Restricted by peopleThat's why I want Google to engage in android, Microsoft to engage in WP, and Nokia to engage in MeeGo at the beginning.
Although HTML 5 is an open system, it is far from mature in the first place, and Facebook's right to speak in standard setting is not big enough. Therefore, according to the final result, there is also a reason to discard it.
However, I think it is not too late to develop mobile applications using Native. For Facebook, the value of its mobile app lies mainly in the content provided by Facebook, not in the interface. At the same time, the mobile platform itself also has ample documentation to illustrate how to develop high-quality Native code, therefore, Native code can be quickly used to bring a better user experience than the HTML5 interface. Html5 solutions for mobile Internet are undoubtedly the most ideal, but the problem is that the current device's support for html5. android2.3 (189/500) and ios5.1 (324/500) are not perfectly supported, under facebook's user base, many problems are inevitable in such a complex environment. Facebook is betting on this. I understand that it is trying to clear these obstacles to build a perfect html5 mobile platform, or to expect users to upgrade their mobile phones faster, but over the past two years, failed to achieve the expected results.
However, this does not mean that html5 is not suitable for mobile development. On ios5.x/ios6, I believe that we can make apps that are not inferior to the native app experience. But more people are still using low-end devices. If the mobile site is not an imitation of native apps, starting from the availability to ensure performance, especially for content-consuming products, it does not require complex interaction. Currently, the html5 solution is completely correct. Now and in the future, web site + mobile site + native apps will form a complete multiscreen ecosystems.
Facebook's engineer blog also said that embracing native applications does not give up html5. many native applications are still made using html5. In the complete Context, Mark adds the following two paragraphs after the sentence is quoted everywhere.
It's not that HTML5 is bad. i'm actually, long-term, really excited about it. one of the things that's interesting is we actually have more people on a daily basis using mobile Web Facebook than we have using our iOS or Android apps combined. so mobile Web is a big thing for us...
We built this internal framework that we called FaceWeb, which was basically this idea that we cocould take the infrastructure that we built out for pushing code every day, not having to submit to an app store, to build Web code on the Web stack that we have, and that we cocould translate that into mobile development. we just never were able to get the quality we wanted...
We can see that:
- The number of Mobile Web apps is still larger than that of native apps, and Mark is still optimistic about the long-term value of HTML5.
- Technically, building Mobile Web apps Based on the FaceWeb platform is far ahead of the current Mobile platform, facebook itself and third-party developers should ensure that the quality of Mobile Web apps in complex environments is difficult enough.
- Based on his acquisition of Instagram, Mark realized that, in the current situation of Facebook, the rich possibilities of native APIs in mobile environments were previously underestimated.
My point of view:
- To target the widest user group, Mobile Web should still be given the highest priority.
- The engineering difficulty of a Mobile Web App with high compatibility is no less than that of a native App
- With the basic service of Mobile Web Components, you must pay attention to native APIs at any time, and be able to find the API nuggets.
Facebook's mistake is that it uses advanced and immature technologies to develop products, resulting in poor product experience.
I chose to use html5 on the mobile platform for several purposes:
1. Rapid Development and iteration;
2. cross-platform compatibility.
But from the current reality:
1. Because of the rapid development of smart terminals and the poor html5 support of various platforms, html5 applications cannot achieve good cross-platform compatibility.
2. The development of wireless networks is seriously unbalanced, resulting in poor area html5 application availability.
3. Because of different rendering methods, html5 has much lower performance than native apps.
In summary, html5 is currently not suitable for mobile platform development. After all the above three problems are solved, html5-based web applications will become a realistic option. Let's talk about my understanding. I want to explain that we are currently doing something similar.
#1. FB also saw the future of the mobile Internet. From the rumors that FB was going to be a mobile phone, we can see that FB attaches great importance to the mobile Internet. Yes, this is indeed the future, but it is not necessarily possible, especially when the old method is used.
#2. There are still many differences between computers and mobile phones. In conclusion, there are more than four or three small computers. Think about how to deal with IE and IMAC in the Internet, but now there is a combination of infinite Screen Lattice and OS version in the mobile Internet. But HTML5 seems to provide the overall solution. So that traditional Internet giants can achieve the same success on mobile without making too much effort. It provides FB with the possibility and temptation to jump from one mountain to another. The result is naturally obvious.
#3. Finally, the relationship between people is expanded through FB, and the world is reduced. However, the Profile of the user that FB is most attractive to external "newcomers" is rather bad on mobile phones. When people come to socialize, they declare their own existence and find the presence of others. If Profile is not more important than relationship, it is at least equally important.
Mobile phones are originally produced by social networking. In the future, FB on mobile phones will be generated. Differences between native and html5:
Go to the cafe, enter for 8 seconds, and bring the coffee up in 3 seconds.
And 3 seconds later.
Which one do you like?
As a bystander, I see the following:
Incorrect growth experience of copying web on mobile
Wrong estimation more mobile devices will be popularized
Wrong time leads to long-term benefits
My experience is that on mobile, it is completely user experience-driven. Mobile design is at least a deep-rooted path.
