To me, at its core the microservices discussion is about explicitly using system and process boundaries as an architectural tool.

Oct 3, 2014 · 6:56 PM UTC

5
36
28
Replying to @stilkov
@stilkov which I think is smart (organizationally) but unfortunate (technically)
2
@pvblivs Yes. In a theoretical, ideal world, this shouldn’t be needed.
Replying to @martiell
@martiell I totally agree, I don’t like the name at all. No chance of that happening though, I fear.
Replying to @stilkov
. @stilkov But system and process boundaries ARE architectural parameters, don't you think?
1
@berkesokhan Yes, depending on what you mean by parameters. I object to them being an implementation detail.
1
1
more replies
Replying to @stilkov
@stilkov @simonbrown Hm… what was client/server about then? N-tier? SOA?
1
@ralfw @simonbrown Fair point. Should have said “architectural tool for domain modeling”, with the obvious connection to a bounded context
@oschoen @ewolff there’s a difference between complicated and complex. The complexity comes, when each team has to master everything
@oschoen @ewolff Though you’re right. The monolith is harder to master. But it’s only “one”
1
@oschoen especially if you have the professional tools, because each of these small problems has to be tackled on every dimension as the big
Load more