exactly, but the point is that we shall not compare apple to oranges. Everyone can develop natively, if he or she wants and needs those features.
It just seems (looking at the market and not only Xojo) that you can’t get everything “natively” if you use x-platform, at least not without many workarounds per targeted platform, and that’s now definitely the road Xojo is taking (at least) in a first step.
only as add-on: my Javas Swing Software has no problems running in terminal Server environments. Sorry for that but it is not native. And also sorry for: I had never any problem running it in such environments.
well the approach they’ve take is very much “lowest common denominator” and anything extra per platform is done with declares etc
HOWEVER, that isnt necessarily the only way to approach things
Its just ONE way - a commonly selected one
There are others
Absolutely. And given that Xojo’s roots are in macOS, I can understand that supporting Windows probably feels a bit like an albatross.
Microsoft has done this kind of thrashing in a few select areas: Windows desktop and remote procedure call frameworks, data access frameworks to a lesser extent, and who can forget the whole classic ASP => original ASP.NET => ASP.NET MVC => ASP.NET MVC Core train wreck, with WASM/Blazor and .NET MAUI web targets now in the mix as well – almost as if they are throwing mud at the side of a barn to see how much will stick. That is one reason I have stuck with mostly back-end work, it has spared me a lot of that turgid nonsense.
In redesigning the system I presided over in my previous gig, I had been using Microsoft Data Access Components (MDAC) technology to read text files and that was completely ditched in Windows 10 and later. Rather than substitute ODBC – itself quite an aging standard and likely to be upended again sometime during the lifetime of my new system – I now read and write text files entirely using System.File.IO classes. I just had to flesh out some existing experimental code with support for joins and for reading fixed-width formats, and now nothing changes unless I change it, lol.
Excuse the delay in my response, I was busy.
Well the writing is on the wall for a while, and always enough customer pressure. It is obvious that MS wants(!) to force classical windows server on Azure connecting via MS Cloud PCs on Azure.
Edit: Reading the article is important, I just saw the preview picture of the link ;-). No MS Office / 365 Apps on a WIndows Server will make it quite useless. But again I had a reason to put a laughing smiley, I think the first rumors, announcements, and corrections go back to early 2010. We will see. MS ist good in making announcements and withdrawing them under customer’s pressure.
your link has nothing to do with RDP/RDS…
I mentioned terminal server only.
And that will happen. But what means that? Nothing. While it has nothing to do with the need of native or not. like @TomasJ mentioned that Flutter and other Web based Component providers having problems I can only say: yes, true. But it is not so with Java Swing and even not with JavaFX in MS Terminal Server environments. So we can take that as evidence that native is not needed for that.
The other way around. You may have even to write two docs: one for Windows, one for Linux. While you have two different UI’s for one Application. And also you have to take care that your App will really run on MacOS and Windows. One Change of one of them and the success Story ends up in a Bug Nirwana.
That is something you have to prevent yourself.
No neither something about Terminal Servers. Firstly the article writes about support of M365 Apps, hence Office and all the rest of crap (Onenote, Teams etc) on Windows Server 2022. Secondly the document is in Preview and lastly, who cares about 2022 in 2026. Until then Microsoft will launch a new server, most probably called 2025.
It’s Microsoft business case, to sell new proprietary versions in shorter terms with so called “added values” nobody really needs.
For many years we (my main customers and I) kicking Microsoft out wherever this is possible. If Microsoft stops Office on Prem forcing everybody to their Azure Subscriptions this would accelerate this development.
back on topic… how hard it’s for x-platform devs to maintain and keep track of different platforms. Unlike thorsten I hold the opinion, that x-platform must always use native controls and design-language of the specific enviroment. X-platform in best case refers to the classes and core functions behind a shiny eye-candy front-end, while the back-end is reused.
Ans as long as I remember x-platform with Xojo was always a struggle with lots of compiler directives regarding these OS specific things.
Nobody wants kinda alien apps in current UI/UX esp. on more advanced systems. How many so called business software still today uses Icons and Toolbars straight from the 1990 with win3.x style? Damn too much!
It’s not safe to assume you speak for everybody.
Modern UIs have been influenced by web designs. If Xojo provided a WebView that enabled you to expose Xojo objects to JS I’d be much more interested in that than any native UI integration they could come up with.
Nobody of the X-Platform Devs is writing what you try to tell here that you provide that. While then you will need one complete Design for every Platform. That is also possible with Java. But I would not do it.
Why? While not only the Controls but also the workflow specifies the platform. MacOS and Windows are totally different. And people like you try to tell in that way: I do it. No. Sorry. Far, really far away from it. While you then have to build two complete different Designs.
And that makes another thing affordable: like you can see there is not the chance for one ui screen design. You have to do all tests for every platform.
Possibly you have such success with your Apps that you can do that. While the platforms having such differences you have to build at the end two different apps.
What I saw until now: the Design Language people like you understood in form of the components. But that is not the Design Language of Windows. I need to laugh when reading that while it has nothing to do with a real platform specific Design. it has only to do with a platform specific component use.
How many of them are written with a native toolkit 
Bright Idea. You may end up with the same kind of UI. That’s the way you do it. Man. Bright
He has his typical Idea. 36 years of Software Business teaching me: no, he is not right. It is something people don’t get. Simply not. I was looking on many Xojo Apps, none of them builded up the correct Workflow for both Designs correctly what is nearly impossible. Design is a Language but it is more than a few components. Exactly that’s the problem. And people do not recognize that.
And so I realized that I have two chances. One is to try to be native on all platforms. Resulted in Xcode for Apple and in C++ for Windows and Linux. Then you have the chance to build up in the correct way.
Way number two: you write a good App with its own UI and concentrate on the most important thing of Design: the usability. And so it ends up. While the described native process of him would break up everything.
Or in other words: a total nice look and feel with the result of a lower App quality is not better.
It is not about IF this will happen, but WHEN.
On April 1st, Microsoft increased its subscription prices by approximately 11% and tied foreign currency prices 1:1 to the USD (This change involves a somewhat opaque method of averaging the exchange rate over a 6-month period, as determined by Microsoft). This “MS exchange rate” will now be adjusted every six months, which may reduce financial planning security for many companies, except the large ones, able and used to hedge against currency fluctuations.
This marks the second significant increase of MS in a short period, accompanied by substantial internal layoffs, yet surprisingly, there has been no decline in subscriptions and little resistance from the market and little coverage by the media.
In Germany, a growing trend that was once unimaginable is now evident, with even DAX companies transitioning to the cloud. Among other reasons, they seek to purchase office services as a commodity mainly to outsource security aspects. They have realized that security risks within their own IT departments are often an issue, as in-house IT personnel may not be up-to-date and can cost more than subscription fees. In addition, buyers and legal departments of companies are often certain that it is easier to negotiate and manage contracts than to trust the expensive internal know-it-alls who recently once again allowed a ransom attack or some other expensive security disaster. Again, that’s not yet true for the whole market, I’m talking about trends.
It is worth repeating that Microsoft has not always been transparent in communicating its strategies (to say it politely). They often test the market and gather feedback before implementing significant changes (“Windows 10 will be our last OS! …”). But anyone who fails to recognize Microsoft’s focus on Azure, subscriptions, and their “cloud first (or only)” approach, including the operating system in the long term, is likely ill-advised.
Of course, this means that some companies are opposed to this trend and against Microsoft’s reprimand, closed-source software in general, and other competing cloud providers. There will be more and more innovative companies that help with this, which is essentially a good thing. However, the trend is clearly moving towards transforming the classical IT landscape into a cloud-based and service-oriented architecture that can quickly and securely connect different (including external) services. For several years, M365 has consistently received the latest features before they are potentially included in the traditional Office package.
For me, there are opportunities in this trend where business IT finally takes center stage, and the core business regains control over the necessary processes instead of becoming helplessly dependent on the incomprehensible and sometimes crazed recommendations of “IT specialists.” The manufacturer that can integrate the most sensible services and, of course, future AI aspects will win the race.
Attractive interfaces and outdated-looking icons will continue to be less of a focus for business decision-makers than the question of whether the technical solution will help their business safely and effectively.
Personally, I am a strong advocate for open source and have faced issues with vendor lock-in situations for decades (likely the most significant problem with Xojo, by the way). However, I have also learned that my personal opinion is typically not very helpful in a business setting if it contradicts prevailing trends.
Very informative post!
I agree with your statements and would like to add the perspective of size of business.
Larger organisations at least have IT departments or a contractor that take care of the IT infrastructure. They also have the necessary resources, knowledge and skills in management and purchasing department for negotiating contracts with IT service providers or vendors.
Smaller businesses who cannot afford these resources are much more dependent on out-of-the-box software.
MS is a master in marketing its products as SME-friendly easy deployment, easy to use.
They also master the art of grooming decision makers. In public services it becomes visible when they develop close ties with MS, in private companies not that much. MS faced antitrust investigations in the US and Europe. Not much has changed since.
It may take some more time for many companies and public services to understand that a lock-in situation is not ideal and can become a threat.
I recently spent a very frustrating morning just trying to get the scripting environment of O365 to reveal itself to me. It was incoming emails I wanted to work with and I ended up setting up a GMail account and just putting a forwarding rule into Outlook.
Thankfully Google’s scripting environment worked exactly as documented.
I am not advocating for Microsoft. But Google is a cloud service too, which is what I meant by “trend.” Their business suite is mature, but they must face the harsh reality that the world is still and sadly not only governed by money but primarily by Excel.
Yes, but about 10 years ago, there were increasingly more small customers who realized that having their own server in the corner wasn’t very smart. Many of them used Windows Terminal Server in external data centers. This trend has even faded in conservative Germany, and now the cloud is no longer taboo. However, nobody is really interested in “cloud” as a term; it’s about purchasing a service, being able to work with it, and reducing the technical aspects to a minimum (especially the operation of the software).
Of course, “trends” implicitly means that, by definition, they don’t apply to every company. Just as one Windows customer has succeeded with Linux, there is also a market for the reverse direction. As is well known, the city of Munich is currently doing this.
LiMux failed, as it often happens, due to poor planning, inefficient project management, and changing political guidelines. However, merely using open source in a thoughtless and deliberate manner leads to no progress either.
NextCloud, for instance, is a great product, but I know many customers who haphazardly run the product on some remote servers without proper security measures, ultimately experiencing failure.
In this respect, it seems logical to leave complex technologies to specialists. Whether it’s Google, Microsoft, Zoho, or whoever is almost secondary. Although many customers simply don’t want to give up their beloved Excel, and certainly not their macros .
In the case of “Nextcloud”: purchase it as a service and refrain from tinkering with it on your own.
I believe the amount of resources that major cloud providers invest in security 24/7 is often underestimated. Internal IT departments, even in large companies, can rarely match this level of commitment and expertise. Admittedly, negative consequences like vendor lock-in remain a significant and important concern, but this issue also exists with a purely open-source approach. Although there may not be a traditional vendor involved, you still find yourself, perhaps even more so, reliant on “specialists.”
And on the subject of costs: in times of corporate behavior policies, license management is becoming increasingly important, from a legal point of view it has always been, but it was often regarded as a trivial offence.
Privately one may hate subscription models, for companies the purchase of services is much easier and the topic of license management is unnecessary because each user is billed automatically. And since “Cash is king” OpEx versus CaPex is playing an increasingly central role, which is becoming increasingly important in corporate decisions, even in smaller companies. Commercially and technically simple upscaling and downscaling is more important than ever, especially for startups.
I’m all for it, both as a user and developer.
Microsoft might be inviting Google to eat their lunch though.
Indeed, only time will tell. It’s certainly a great opportunity for businesses that can move beyond the notion that MS Office is the epicenter of the world and that everything would collapse without it.
Even though every financial controller and other bean counter in large corporations might want to strangle me on the spot for saying this, I stand by my statement
I’d prefer not to comment on the quality of Microsoft’s documentation. Specifically, in this country, one frequently encounters Microsoft’s automated translations as the primary source. While these translations can be amusing, they are often more unhelpful than not.