So, then, given that the PgAdmin 4 lineage just by itself is now 13 years old
and a lot has changed since then in the technology landscape...
Hypothetically, would you consider starting a PgAdmin 5 written in Java or some
other stack, or would you still choose the same stack you have now to make a
major new version today?
For me personally, by coincidence, Java is my top preferred and recommended
application language, when the problem domain does not specifically favor
something else, and I feel Java continues to receive vast useful improvements
each year. Key factors include it having strong compile-time checking and typing
while still heavily supporting generics, and having multiple major vendors
supporting the core product.
Despite that, I do not currently feel that PgAdmin 4 is not "modern" or needing
a rewrite for technology sake.
And I agree with the core design decision of a common code base for both a
desktop application and a web application, this helps with the longevity of the
current version, although at the same time also making it more structurally
adaptable to piecewise rewrite, such as replacing the Python layer with Java
while keeping the Javascript layer as that, not that there are really any
alternatives to running in a web browser that aren't involving Javascript.
On 2026-10-13 6:48 a.m., Dave Page wrote:
You're not wrong, but at this point I wouldn't use the word "new" for pgAdmin 4
:-)
Screenshot 2026-10-13 at 14.45.49.png
pgAdmin 1 (well, just pgAdmin back then), actually started in 1998/1999. I'm not
sure of the exact date - the predecessor, pgManager was never released and that
only lived for a brief while in 1998 or so. For those that remember it, it was a
Visual Basic 4 clone of the original PostgreSQL GUI, pgAccess, written in TCL/TK.
On Tue, 13 Oct 2026 at 14:31, Darren Duncan wrote:
In the scheme of things, PgAdmin 4 is already an ongoing rewrite in a new
language, Python, compared to PgAdmin 3 that was written in C++.
It seems like the team already made their language choice for the "modern"
version of the tool.
I would be surprised if they turn around and rewrite it again in Java any
time
soon, since the Python one is relatively new.
Granted, with the new LLM tools we have now, choosing to rewrite a codebase
from
one language to another isn't necessarily as onerous or time consuming as it
once was.
What is the business case to do it now though?
On 2026-10-12 2:19 p.m., Rick Yost NC Data Team wrote:
> Are there any plans to convert PGAdmin to Java? I deal with very large
databases
> and some of the queries I run produce millions of records, I've been
forced to
> use other products to extract the data and run my queries, but I always
like
> pgAdmin4 for the smaller stuff. I wish it would do it for the bigger
stuff. But
> since it's based on chromium, apparently there's a problem with
downloading that
> many records after I run the query.
