VMware published a blog “Goodbye vSphere Client for Windows (C#) – Hello HTML5” with two announcements: 1. There won’t be the traditional C# based vSphere Client for the next version of vSphere (no one knows the version yet but it’s safe to call it the next version.). 2. The vSphere HTML5 Web Client Fling will be supported soon (no exact timetable is given). My guess is that it will make into next release of vSphere if not earlier.
To show VMware’s strong commitment, the new HTML5 Client will be referred as vSphere Client moving forward. Wouldn’t that cause confusion? I personally think it would and thus doubt it will happen. I believe most people will simply refer it as vSphere HTML5 Client, not as mouthful as the “vSphere HTML5 Web Client” mentioned in the blog.
Lost VMs or Containers? Too Many Consoles? Too Slow GUI? Time to learn how to "Google" and manage your VMware and clouds in a fast and secure HTML5 App.
As much as people like the idea that VMware will support the HTML5 Fling, people hate the idea that VMware will take away the C# based vSphere Client. Right after the announcements, there are many customers left comments (more than 80 comments as of this writing) questioning VMware’s decision, mostly the discontinuing of the C# client.
In fact, this is not the first time that VMware wanted to kill the C# based client. It announced a couple of times already since the vSphere 5.5, and then took back due to too strong oppositions from customers who found the replacement vSphere Web Client is too slow to be useful, not to mention it also lacks features compared with the vSphere Client.
Fundamentally, I think VMware needs a standalone client as its main GUI for vSphere. Although Web applications are getting more powerful, but they still lack some features as I discussed 4 years ago in my blog post. To make up the shortcoming of the Web, VMware has something called Client Integration Package that well defeats the original claimed benefit of using Web. Even as today, I am still amazed that in the mobile world everyone talks about (native) apps while in the desktop world, people like more Web applications. Simply because of screen size difference?
That doesn’t mean the Web should not be used as front end for vSphere. At doublecloud.net, we actually pioneered in creating Web application that help manage large scale of vSphere deployments. In the coming release of vSearch, we will add significant features toward our “super vCenter” vision.
While many customers have given good reasons for VMware not to discontinue the C# vSphere Client from their business perspectives, I’d like to offer a few reasons from technical perspective. (I haven’t read through all the comments, so if some of reasons are already raised up in the comments it is purely coincidental).
First, and foremost, is the blog I wrote before, I think Web application (either Flash based or HTML5 based) is nice but it’s not good enough as the main vSphere GUI.
Second, it takes time for a product to mature. It’s nice for the Fling to be well received by the trial customers, but the expectations would be much higher when a product is used in production environment as main GUI.
Third, the C# client and the HTML5 client are actually complementing each other. The real competing product VMware should say good bye to is the vSphere Web Client.
To be realistic, I think the reasonable actions VMware should take would be:
1. Continue the vSphere Client. Even nice if VMware can support new features into the product.
2. Discontinue the vSphere Web Client and replace it with the HTML5 Client. Because there are quite a lot of customers having invested a lot in creating plugins for the vSphere Web Client, VMware should help them to make a smooth transition. Having one release window for partners should be a reasonable expectation.