The ISRLN website says nothing about who is behind it. It in fact says nothing about who is behind it so loudly as to cast the whole enterprise into doubt.
Wikipedia and the European Language Resources Network pages in ISRLN do given an answer for at least the first question. The 12 organisations that got together in Paris in 2013, to establish the ISRLN, are:
Alliance of Digital Humanities Organizations (ADHO), Association for Computational Linguistics (ACL), Asian Federation of Natural Language Processing (AFNLP), COLING Committee (ICCL), European Data Forum, European Language Resources Association (ELRA), International Association for Machine Translation (IAMT), International Committee for the Coordination & Standardisation of Speech Databases and Assessment Techniques (COCOSDA), International Speech Communication Association (ISCA), Linguistic Data Consortium (LDC), Oriental COCOSDA, Language Resource Management Agency (RMA)
ACL, COLING, and the LDC are real enough to encourage some confidence. The portal was "Set up by ELRA, LDC and AFNLP/Oriental-COCOSDA".
The ISRLN is being used in the ELRA's catalogue of language resources: http://catalogue.elra.info/en-us/repository/browse/training-and-test-data-for-arabizi-detection-and-transliteration/e2080666699611e8886702b3f1438bcb8e338a7b921a4a858d7dc4be3b92580d/ => http://www.islrn.org/resources/986-364-744-303-9/ ; so there is dogfooding.
Questions of persistence for Yet Another Persistent Identifier are always appropriate, and I'm astonished myself that they didn't just go with DOI as an identifier. But notwithstanding how uninformative the ISRLN website is, the enterprise is for real, and does have some institutional backing.