DECIPHERMENT OF LINEAR B
Early attempts at deciphering Linear B were all based on the assumption that it transcribed an unknown language used only by the Cretans who built Knossos. However, Ventris believed that the language might have been something based off of Greek or Etruscan. This theory was supported by the research of Alice Kober, who noticed that similar strings of characters would often have varied endings, which suggested that there were conjugations, as in Latin. Ventris tried many means to break Linear B but the two methods that proved most useful were designing a syllable grid and using frequency analysis.

When building his syllable grid, Ventris tried to figure out which symbols where related to each other, either by having the same consonant or the same vowel sound. Finding relationships between consonants proved the easiest because in a language that is conjugated, often the endings differ only in the vowel but the consonant is the same. An example of this is in Latin. The noun dominus would have the following forms: dominus, domini, domino, dominum. All four forms end with an "n" sound (-nus,-ni,-no,-num). If Latin was written with a syllabary, all four endings of dominus would appear together in the syllable grid.
sample syallable grid

When using frequency analysis, he realized that symbols that represent stand alone vowel sounds ("a", "e", etc.) where more likely to occur at the beginning of words than at the middle or end. Using this idea, he was able to guess at symbols most likely to be vowels.

With some guesses of what each symbol could have stood for, Ventris started playing around with the texts, attempting to find the names of Cretan cities, such as Amnisos (which he assumed would transcribe as a-mi-ni-sos). He had a guess for what was "a" and what was "ni", so he looked for places in the tablets where these two symbols appeared in the right order. They only appeared like that once. That gave him a reasonable guess for what could be "so" and "mi". Next, he tried looking for Knossos (written ko-no-so). With a value for "no" in his chart and a possible value for "so" coming from his Amnisos search, he found a possible match for Knossos.

As he went through the texts, trying to find matches for the symbols, he came across combinations of syllables which matched with classical Greek words. The first hint was various place name endings which matched those used in Greek. Then, he became to notice actual Greek words in the text. Finally, after testing the texts to his satisfaction, he came to the conclusion that the language transcribed in Linear B was indeed Greek. With that discovery, he was able to say that Linear B had finally been broken.