Distributed software library

We propose the design and usage of a distributed Directory Service. This paper also introduces our hybrid approach using two networks and a combination of different distributed routing strategies for event filtering. Documents: Advanced Search Include Citations.

Authors: Advanced Search Include Citations. Keyphrases open digital library software distributed alerting service hybrid approach digital library open source software independent service distributed digital library server actual running system distributed digital library distributed directory service event filtering digital library system useful feature library user constitute so-called metasoftware new way routing strategy fragmented network dl software publisher-hosted proprietary service.

That way moving forward is faster than dealing with the contribution to the shared libraries which often involves discussions about who should be able to contribute to the library project, etc. An additional reason for not creating shared libraries is that these are not so portable, at the end of the day. Knowing these limitations, it may seem clever to just document in detail the specification of what the library brings to the client applications, with the code samples from the existing apps.

That way the authors of the same functionality for the other programming language just need to read the specification, and implement the solution and document it the same way for all the existing languages. In the world of distributed software, and having multiple development teams, the strategy to avoid shared libraries may sound like a sane strategy to deal with various business requirements that affect of codebases of multiple teams.

Finding the right group of engineers to implement the initial solution and document it well is the most important step. After that, each development team needs to be able to understand the scope of the functionality and to be able to integrate the solution into its own services and eventually - to be able to contribute to any enhancements needed. Shared libraries in distributed systems Date published About shared libraries Shared libraries could be, in simple words, defined as a piece of code, often developed by somebody else, and that we can use in a relatively straightforward way.

I would like to make a distinction between two kinds of libraries: Publicly available libraries In house libraries Publicly available libraries The open-source community nowadays provides a huge number of useful libraries that are generic and address common issues, that usually appear while developing software nowadays. In house libraries The main focus of this writing is the other kind of libraries that are present as well, especially in mid-size or big size companies.

What could be the domain of such libraries? Fulfilling legal requirements when using some technology There are some cases that some legal requirements or special contracts with some clients force generation of some kind of shared libraries, that seem to address these requirements the best way there is. Shared libraries are not portable everywhere An additional reason for not creating shared libraries is that these are not so portable, at the end of the day.

Also, investing in them is not portable to the other languages that may exist in the system. Conclusion In the world of distributed software, and having multiple development teams, the strategy to avoid shared libraries may sound like a sane strategy to deal with various business requirements that affect of codebases of multiple teams.

That was all for today! Hope you liked it! Do you have a question or comment? Glad to discuss it with you on Twitter.