The reason seems basically external: Gandalf was a wizard in The Hobbit, but not yet an angelic being sent by the Powers, and I guess Elrond actually knew more about some runes than Gandalf.

The sword-names are in Sindarin (actually there was no 'Sindarin' when Tolkien wrote The Hobbit).

In the 1960 Hobbit, meant as a revision of the tale so that it would fall in line better with The Lord of the Rings, Tolkien basically rewrote this part so that Gandalf could not read the runes (the runes on the swords) because there was dried blood obscuring the writing.

Certain websites suggest that the runes were 'Gondolinic runes' -- and thus better known to Elrond than to Gandalf -- but we see that Tolkien rather came up with dried blood, although in any case he never introduced this into a later, revised edition of The Hobbit.

It gets complicated and tricky, but when Tolkien wrote The Hobbit the scenario with respect to the Elvish languages of Middle-earth was very different from what it would later become.

Rather than explain that confusion, basically the sword-names Orcrist and Glamdring 'became' Sindarin after 'Sindarin' was invented, and after drastic changes in the internal history of the Elvish languages came about, which were imagined quite late in the making of The Lord of the Rings and appendices.

Thus we adopt these names into the ultimate scenario, and assume that, generally speaking, since Sindarin was spoken in Gondolin (Quenya was spoken in Turgon's house if I recall correctly), the sword-names were simply written in Sindarin -- perhaps in the Angerthas Daeron writing system (chosen by the filmmakers), or whatever writing system was employed in Gondolin.

This isn't that odd, since Turgon, Glorfindel, Ecthelion are all Sindarin names, for examples. It is arguably odd that Gandalf couldn't read the writing; and as we see, Tolkien thought so too.