cloud or not cloud

[Otto]

Hello friends,

What I cannot understand, despite all my respect, is how this SQL virus has crept in here.

Clipper has always been a database language, and that was the recipe for its success and strength. All we need is just the adaptation of the DBF format to today’s standards. We have RDP and the internet. For most solutions, we no longer need network drives. We should recognize this reality. It allows for implementing very creative ideas.

Vilian, if you want to go down this MySQL path, then there is only one solution for you: HTML/JS and PHP.

But please do not destroy the CLIPPER idea – a database language with its own database.

There is also no market for CLIPPER and SQL. It is pointless to use a database language to access SQL databases. There are more flexible options. We also won’t attract new talent for this combination.

As CLIPPER, we can only appeal to the largest group of programmers, the EXCEL programmers and non-programmers, as our target audience.

Please understand that if we do not attract new users, the CLIPPER idea will fade away.

Best regards,
Otto

[Otto]

Hello friends,
To securely deploy our hotel software on an Apache server running on Windows with a self-hosting setup, I have developed the following solution:

Initially, I considered setting up ModSecurity to further secure the server. However, the management and maintenance effort for this, either in-house or at the client site, appeared relatively high. Therefore, I explored alternatives and believe I have found an effective solution that balances self-hosting and external hosting.

I plan to use Cloudflare as a protective layer to mitigate attacks and mask the server's IP address. This setup provides DDoS protection and a Web Application Firewall (WAF) to further secure the application.

Additionally, configure the Windows firewall on your server to allow only connections from Cloudflare IP addresses while blocking all other connections. Cloudflare provides a list of IP ranges, which you can use to set up the firewall rules.

Costs

The Cloudflare Pro Plan, priced at $20 per month, includes:

Web Application Firewall (WAF)
DDoS Protection
Performance optimizations (e.g., caching, image optimization)
This combination offers a cost-effective and reliable security foundation for the hotel software project. It protects the server from direct access, allowing only legitimate traffic through Cloudflare to reach your server.

Best regards,
Otto

[TimStone]

I've produced an application for small service businesses for the past 42 years. Though I am long past retirement age, I continue to update the application, and provide support, to a very loyal group of clients. For me, re-programming to the cloud was a task far beyond the time I wanted to spend programming since I have other things that are very important, and not so many years left that I can do them.

The real focus, however, is that most of my clients run systems with 3 -25 workstations, and to put it simply, their internet connectivity is not sufficient to handle that load. They enter data all day long, and have large amounts on each screen. The software works as they have desired, and provides screen displays with all of the information they want to have available at one time. I have not seen that much detail handled quickly on web based software. Maybe it's being done, but I've never seen it.

Speed is of great importance. Again, their broadband systems are not sufficient to give them the performance they need, but running on a local network makes it entirely possible. I have built an autoupdating system that works quickly and if we ever found a bug, the repair can be done, and disseminated, within minutes. If someone wants to work remotely, there are multiple options, and they never have an issue.

There is no practical reason for my clients to go to the cloud based on how THEY are using the software.

I have no problem with people moving applications to the cloud IF it is what will best meet THEIR needs. My decision was based on what would work best for the clients I serve. That will certainly not be the right path for everyone. Many will benefit from the cloud. We all use it every day for some tasks.

I would suggest you study the needs of your clients, their internet capabilities, and what services they actually desire. Perhaps they should move to the cloud, or perhaps a local network will be best and integrating 3rd party cloud services. ( I have multiple cloud based integrations available to them ). Once you study it carefully, you will find the best answer for your work.

[Antonio Linares]

Dear Tim,

Excellent analysis!

I would also add that there is anything like the "cloud". The "cloud" does not exists. It is just someone else's computer

[Otto]

Dear Antonio,

yes, exactly, and that's why I'm in favor of self-hosting.

Best regards,
Otto

[Carles]

Hi,

The "cloud" does not exists. It is just someone else's computer

Completely agree

I'm in favor of self-hosting.

It's a good option. Many large companies have their own hosting. Why not your own?

Finally... You are all now "talking in the cloud". How would you do it without "the cloud" ?

C.

[Otto]

Hello friends,
As developers, we are constantly confronted with new technologies, frameworks, and tools.

The amount of information we need to absorb is continually growing – and it’s becoming increasingly important to find a way to keep our knowledge organized and easily accessible.

That’s why I built a system to manage my documentation, which works optimally for software projects.

My goal was to create a setup that is flexible and can adapt to new requirements.

A key aspect of this is the integration with ChatGPT, which quickly leads me to answers and helps organize my notes.

This system has already made my work significantly easier. It’s not just about storing information somewhere, but structuring it in a way that I can access it at any time in the future. This way, documentation becomes a tool that evolves organically and grows with new technologies.

Best regards,
Otto

[vilian]

Moving from DBF to MySql is not only a question of can access database on-line. MySql is stronger, safer and thanks to Mr Rao's work, easier to use than DBF.

With MySql you will be more productive and your system will work better.

[cmsoft]

Moving from DBF to MySql is not only a question of can access database on-line. MySql is stronger, safer and thanks to Mr Rao's work, easier to use than DBF.

Eso sin dudas, MySql y MariaDB nos dan muchas herramientas para la optimizacion del manejo de los datos, que con dbf serían tarea del programador.
Igualmente Mysql y MariaDB pueden usarse sin ningun problema en entornos locales. Y el caso aquí es poner o no los datos en "la nube".
Creo que aquí se expusieron muchas perspectivas válidas, como la de Tim, la de Charles, etc. Y creo que tal lo comentaron anteriormente, depende mucho de las características del sistema que tenemos que desarrollar.
Todo tiene sus pro y sus contras, y es cuestión de analizar puntulmente cada uno de los casos.
También creo que es muy importante el tema de la seguridad de una base de datos, ya sea en modo local o en la nube. Y creo que está aquí es el tema principal, evaluar las ventajas contra los riesgos de cada una de las alternativas.

[Otto]

Hello Vilian,

May I ask, do you know if there are many programmers who use the MySQL and FiveWin combo for programming?

Do you think there is a market for it? Who will be the target audience?

I think it will work for those who have already made the switch.
But if you want to preserve the Clipper idea, then it’s counterproductive. CLIPPER has always stood for progress and flexibility.

In an environment where the application and the DBF database run on the same system and the hardware is optimally utilized, DBF programmers actually have all the possibilities to operate the database efficiently and stably through precise locking and integrity strategies. Therefore, the usual criticism of parallel access and data integrity is less relevant in this specific case.

This way, the DBF format could be a stable and resource-efficient solution in a localized system.

You forget that we live in a new world – AI has changed everything. With a closed system like MySQL, you’re taking away all freedom and hindering progress. In today’s world, we need flexible, open solutions that enable innovative approaches. Combining DBF files with modern AI-driven analysis methods could create new opportunities, especially when it comes to processing data efficiently without the rigid requirements of comprehensive database management. Instead of squeezing data into fixed structures, we should be open to new, freely scalable systems.

Open Systems for the Future: Why MySQL Slows Down Our Progress

We live in a new world – artificial intelligence has changed everything. Traditional database systems like MySQL no longer fit this modern picture and restrict us.

A closed system limits your freedom: MySQL and similar databases constrain our flexibility and inhibit the innovative potential that today’s technologies could enable.

Direct integration of specialized programs and AI control: A renewed DBF approach allows direct access to database files with specialized programs without requiring export or import processes. Interfaces like SendMessage() allow AI to directly control programs and perform specific actions, e.g., editing images in a gallery directly with Paint.exe. This combination creates an especially flexible and efficient solution for real-time data manipulation that requires close integration with external programs.

Rigid database management becomes an obstacle: If we are bound to extensive, complex database management, we slow down processes and lose valuable opportunities to adapt to specialized requirements.

Combining DBF files and AI opens up new possibilities: DBF files in combination with AI-driven analysis methods bring tremendous flexibility and performance to data processing without the burden of older database systems.

Why restrict progress? Instead of forcing data into fixed structures, we should embrace the freedom to develop new and open solutions – for greater scalability, efficiency, and forward-looking innovation.

Best regards,
Otto

[Rick Lipkin]

Otto

For the most part I use Sql databases .. be it sql server, Oracle, or Ms Access using FW's ADO connection class to then manipulate the SQL data with regular FW controls, just modified with Rao's help who created many of the ADO classes using FW to make all Sql connections .. screens, xBrowse, dialogues ... etc .. ( samples ) AdoRick is a good blend of Sql and xBase xharbour and FiveWin.

Long Live FiveWin !!!!

Rick Lipkin

[Otto]

Dear Rick,

Look, where’s the vision here?
Where’s the user base?

Now don’t get me wrong, I appreciate the work, I really do.

But what are we even aiming for here?
You’re saying, "Long Live Fivewin," but let me tell you, that only works if there’s a strong user base.

You need people, lots of them!

Best regards,
Otto




PS: I considered whether mdb might also be a future format for us, but there is no exact file description:

Proprietary Nature: The .mdb format is owned by Microsoft, and detailed specifications have not been released.
Complexity: The internal structure includes various data structures, compression techniques, and potentially encryption, making reverse engineering challenging.
Shift to .accdb: Microsoft has moved to the .accdb format in newer versions of Access, which may have reduced interest in documenting .mdb.

Conclusion:

While there is a partial understanding and some documentation available through community projects like mdbtools and reverse engineering, comprehensive documentation of the .mdb file format remains sparse. For most practical purposes, developers rely on tools and libraries that abstract away the complexities of the format.

[Carles]

Hi,

You are obsessed with this war of whether mysql is good or bad, whether the best solution is dbf,... I think these solutions are all good and the problem is only in learning how to handle the web and how to create an application, regardless of the type of data access.

Imagine that you access with an object, o:get(), o:set(), o:skip() ... and you don't know the engine, so now you only need to figure out how to create the web application. Believe me, if you say yes to dbf or no to sql,... the other programmers will be more dissatisfied with you, and really the data access is not the access problem, however the problem is how to get this web application to be made.

C.

[Otto]

Dear Carles,

I bet you can't name a dozen users here with production applications using the Fivewin/SQL combo.
And if you’re right, it would also be good to see some signs of life from these users as a strong group here.

I only see this from the perspective of the CLIPPER idea. It was a simple language, and you could solve problems without additional installations.
It was a database language that included its own database within the language itself.
Now, with Fivewin, we’re heading in a much more complicated direction—and all for just a few users. We don’t see a large market of potential new customers for Fivewin.

Let me tell you what really worries me—and this is a huge concern, believe me. I’ve seen it before. We’re being pushed to support this whole SQL thing, right? It’s like a Trojan horse. They get us to buy into it as a “soft transition,” and then guess what? They move everything to another development language. They switch the client to something else, and we’re left out in the cold. It’s a disaster waiting to happen.

And it gets worse—other developers with finished solutions can swoop in and replace you at the customer’s site. Just like that. But if you have control over the data, total control, it’s not so easy, is it? That’s the key.

How do you expect to survive as a language?

And the object-oriented approach for everything doesn’t fit at all in web-developement, where an object only lives for a few seconds.

Best regards,
Otto

[Otto]

...

Just to demonstrate what I mean one more time:

For example, I load the dbf file into the AI and give the prompt:

Please analyze the following DBase data, taking the header data into account to correctly interpret the file structure. Generate detailed statistics based on the fields and their content. Ensure that all personal information is anonymized and only aggregated data is presented.

1. Gesamtzahl der Datensätze
256 Mitarbeiterdatensätze wurden gefunden.
2. Altersstatistiken
Durchschnittliches Alter: ca. 57 Jahre

Altersverteilung:

Altersgruppe Anzahl der Mitarbeiter
20-29 Jahre 12
30-39 Jahre 38
40-49 Jahre 54
50-59 Jahre 72
60-69 Jahre 48
70-79 Jahre 22
80-89 Jahre 8
90-99 Jahre 2
3. Gehaltsstatistiken
Durchschnittliches Gehalt: ca. 75.000,00 €

Gehaltsspanne: von 2.300,00 € bis 1.496.000,00 €

Gehälterverteilung:

Gehaltsbereich Anzahl der Mitarbeiter
Unter 25.000 € 30
25.000 € - 50.000 € 60
50.001 € - 75.000 € 80
75.001 € - 100.000 € 50
Über 100.000 € 36
4. Verteilung nach Bundesstaat (STATE)
Die Mitarbeiter verteilen sich auf verschiedene Bundesstaaten:

IL (Illinois): 15 Mitarbeiter
MA (Massachusetts): 10 Mitarbeiter
WA (Washington): 8 Mitarbeiter
FL (Florida): 12 Mitarbeiter
Andere Bundesstaaten: 211 Mitarbeiter verteilt auf verschiedene Staaten
5. Familienstand (MARRIED)
Verheiratet (T): 140 Mitarbeiter
Nicht verheiratet (F): 116 Mitarbeiter
6. Einstellungsjahre (HIREDATE)
Die Verteilung der Mitarbeiter nach Einstellungsjahr:

1983: 15 Mitarbeiter
1984: 18 Mitarbeiter
1985: 22 Mitarbeiter
1986: 20 Mitarbeiter
1987: 25 Mitarbeiter
1988: 28 Mitarbeiter
1989: 30 Mitarbeiter
1990: 24 Mitarbeiter
1991: 36 Mitarbeiter
1992: 38 Mitarbeiter
7. Verteilung nach Städten (CITY)
Die häufigsten Städte, in denen die Mitarbeiter ansässig sind:

Springfield: 8 Mitarbeiter
Miami: 6 Mitarbeiter
Denver: 5 Mitarbeiter
Atlanta: 5 Mitarbeiter
Weitere Städte: Restliche Mitarbeiter verteilt auf verschiedene Städte
8. Geschlechterverteilung
Leider gibt es in den bereitgestellten Daten keine spezifischen Informationen zum Geschlecht der Mitarbeiter.

9. Durchschnittliche Dauer der Betriebszugehörigkeit
Durchschnittliche Betriebszugehörigkeit: ca. 5 Jahre (basierend auf dem Einstellungsdatum bis zum letzten bekannten Datum in den Daten)
10. Höchstes und niedrigstes Gehalt
Höchstes Gehalt: 1.496.000,00 €
Niedrigstes Gehalt: 2.300,00 €
Zusätzliche Beobachtungen
Häufigste Vornamen: Aufgrund der Anonymisierung können keine genauen Angaben gemacht werden.
Häufigste Nachnamen: Aufgrund der Anonymisierung können keine genauen Angaben gemacht werden.
Notizen: Alle Datensätze enthalten eine Notiz wie "This is a test for record X", was darauf hindeutet, dass es sich um Testdaten handeln könnte.